- Email: [email protected]
Preservice teachers’ implicit attitudes toward racial minority students: Evidence from three implicit measures
Studies in Educational Evaluation 45 (2015) 55–61
Contents lists available at ScienceDirect
Studies in Educational Evaluation journal homepage: www.elsevier.com/stueduc
Preservice teachers’ implicit attitudes toward racial minority students: Evidence from three implicit measures Sabine Glock a,*, Julia Karbach b a b
Bergische Universita¨t Wuppertal, Germany Goethe University Frankfurt, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 June 2014 Received in revised form 16 March 2015 Accepted 17 March 2015 Available online 11 April 2015
Implicit attitudes can be activated by the mere presence of the attitude object. They are assumed to guide behavior in demanding situations, including the educational context. Implicit attitudes toward racial minority students could be important in contributing to the disadvantages those students experience in school. This study employed three different measures to investigate implicit attitudes toward racial minority students among preservice teachers. The IAT and the AMP showed more negative implicit attitudes toward racial minority than toward racial majority students; the affective priming task revealed that implicit attitudes toward racial majority students were positive, while those toward racial minority students were neutral. Results are discussed in their implications for preservice teachers’ judgments and behaviors. ß 2015 Elsevier Ltd. All rights reserved.
Keywords: Racial minority students Implicit attitudes Implicit measures Preservice teachers
Introduction Educational research has consistently shown that racial minority students experience disadvantages in educational systems. Compared to racial majority students, they perform poorly in school (e.g., Dee, 2005; Haycock, 2001; Lee, 2002) and thus are overrepresented in lower level and vocational courses (Ansalone, 2001; Lucas, 2001; Oakes, 2005), in lower school tracks (Caro, Lenkeit, Lehmann, & Schwippert, 2009). They are also recommended less frequently for the higher level tracks, even when academic performance is controlled for (Glock, KrolakSchwerdt, Klapproth, & Bo¨hmer, 2013). In Germany, racial minority students mainly stem from Turkey (Destatis, 2012). Students with Turkish roots are overrepresented on the lowest school track (Caro et al., 2009) and underrepresented on the highest track (Kristen & Granato, 2007). They more frequently fail to complete school (Coneus, Gernandt, & Saam, 2009) and consequently, they have jobs with lower prestige and lower employment rates than their German peers (Euwals, Dagevos, Gijsberts, & Roodenburg, 2007). In the German school system, attending the highest school track and leaving this track
* Corresponding author at: Bergische Universita¨t Wuppertal, School of Education, Gaußstraße 20, 42119 Wuppertal, Germany. Tel.: +49 0202 439 3082. E-mail address: [email protected] (S. Glock). http://dx.doi.org/10.1016/j.stueduc.2015.03.006 0191-491X/ß 2015 Elsevier Ltd. All rights reserved.
with a qualification for university entrance is of high importance for the future professional career of students. Considering that teachers are the main decision makers when it comes to grading or tracking (Ansalone & Biafora, 2004), the abovementioned disadvantages might not only stem from racial minority students’ lower academic achievement, but also from biases in teachers’ judgments. In educational research, stereotypes and teacher expectations are discussed as factors influencing teachers’ judgments (e.g., Jussim & Harber, 2005; Su¨dkamp, Kaiser, & Mo¨ller, 2012). To this extent, teachers expect racial minority students to show lower academic achievement than racial majority students (Tenenbaum & Ruck, 2007). Teacher judgments have been shown to be affected by race (McCombs & Gay, 1988; Parks & Kennedy, 2007), indicating a possible negative bias against the racial background of a student; this was true for both experienced and preservice teachers with limited teaching experience. In this vein, negative teacher and preservice teacher biases might reflect prejudice (Devine, 1989), defined as negative attitudes toward the members of a social group (Dovidio, Brigham, Johnson, & Gaertner, 1996). Attitudes reflect the positive or negative evaluation associated with an object or a social group (Fazio, 2007) and attitudes might affect how people are perceived and judged (Olson & Fazio, 2009; Sanbonmatsu & Fazio, 1990). Thus, it seems of high importance to investigate the nature of attitudes toward racial minority students, as those attitudes might be reflected in judgments. Hence, the aim of this study was to
56
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
investigate preservice teachers’ attitudes toward racial minority students.
Theoretical background On a theoretical level, implicit and explicit attitudes are distinct constructs resulting from two different mental processes (Gawronski, Strack, & Bodenhausen, 2009). Explicit attitudes are defined as conscious evaluations of the attitude object which result from reflected and controlled processes (Gawronski & Bodenhausen, 2006). In contrast, implicit attitudes are defined as automatic evaluations of the attitude object which result from automatic processes (Gawronski & Bodenhausen, 2006). Implicit attitudes can be activated by the presence of the attitude object (Fazio, 2001). Thus, the different kinds of attitudes resulting from different mental processes are prevalent in different situations. These processes are specified in dual process models on how attitudes guide behaviors (e.g., Fazio, 1990; Gawronski & Bodenhausen, 2006; Strack & Deutsch, 2004; Wilson, Lindsey, & Schooler, 2000). In this study, we draw on the Motivation and Opportunity as Determinants (MODE) model (Fazio, 1990; Olson & Fazio, 2009). Explicit attitudes are assumed to guide perception, judgments, and behavior in situations where people have enough time, cognitive resources, and motivation to reflect on their attitudes, make conscious judgments, and control their behavior (Fazio, 1990; Olson & Fazio, 2009). In situations with high cognitive demands, implicit attitudes should be prevalent and guide perception, judgments, and behavior. Although these two processes are not mutually exclusive and both explicit as well as implicit attitudes might come into play (Olson & Fazio, 2009), the automatic character of implicit attitudes means that implicit attitudes are dominant in most situations, as they are automatically activated by the mere presence of the attitude object (Fazio, 2001) and people are often not aware of this influence (Asendorpf, Banse, & Mu¨cke, 2002). Implicit and explicit attitudes are measured using quite distinct methods. Explicit attitudes measures rely on the assumption that explicit attitudes are conscious evaluations and that people can report their explicit attitudes when asked to evaluate an object. Thus, explicit attitudes are assessed using questionnaires and rely on self-report measures. However, when it comes to socially sensitive issues, people are often reluctant to report their ‘‘real’’ explicit attitudes (Dovidio & Fazio, 1992; Fazio, Jackson, Dunton, & Williams, 1995), and explicit attitudes measures might reflect social norms (Fazio et al., 1995; Karpinski & Hilton, 2001) and social desirability (De Houwer, 2006) rather than explicit attitudes. Hence, implicit attitudes measures attempt to reduce the social desirability bias (De Houwer, 2006). Since implicit attitudes are defined as being automatic, implicit attitudes measures tap into automaticity (De Houwer, 2006; Moors & De Houwer, 2006). The definition of automatic processes involves unconsciousness, nonintention, unawareness, and efficiency. Therefore, implicit measures usually share at least one of these properties (Hofmann, Gschwendner, Nosek, & Schmitt, 2005). As a whole, implicit attitudes measures seem to be a valuable tool in assessing socially sensitive issues such as attitudes toward racial minority students. Implicit attitudes may also play a pivotal role for preservice teachers’ judgments and behaviors, since working as a teacher is stressful (Van Dick & Wagner, 2001), often requiring action under time pressure (Santavirta, Solovieva, & Theorell, 2007). Under such conditions, implicit attitudes may be particularly influential. Given that preservice teachers’ judgments can have a great impact on students’ future educational and professional careers, preservice teachers’ implicit attitudes toward racial minority students are vital when they enter the classroom.
In the last years, many different implicit measures have been developed (see Glock & Kovacs, 2013, for an overview of most but not all measures). Since previous research on implicit attitudes has shown that the nature of implicit attitudes differs as a function of the stimuli used (Robinson, Meier, Zetocha, & McCaul, 2005) and the methods applied (Sherman, Rose, Koch, Presson, & Chassin, 2003), inconsistent results regarding implicit attitudes toward racial minority students could stem from those measurement methods. Although research employing the same three implicit measures has not found substantial differences in revealed implicit attitudes, it is nevertheless plausible that each measure assesses unique aspects of implicit attitudes (Payne, Govorun, & Arbuckle, 2008). Moreover, each implicit measure requires different categorization tasks (Olson & Fazio, 2003) and participants’ performance might not only reflect automatic attitudes, but also particular features of the stimuli or of the categories employed in the measure (De Houwer, Geldof, De Bruycker, & De Bruycher, 2005; De Houwer, 2003). Hence, employing multiple measures of implicit attitudes is highly recommended in implicit attitudes research. Therefore, in order to rule out that measurement methods affect the results to a greater extent than implicit attitudes do, we employed three different implicit attitudes measures using the same stimulus materials. Research question The aim of this study was the investigation of the nature of implicit attitudes toward racial minority students. Educational research on implicit attitudes toward racial minority students is particularly sparse. One study found ambivalent implicit attitudes toward racial minority students among preservice teachers, while the attitudes toward racial majority students were positive (Glock, Kneer, & Kovacs, 2013). Another study found slightly more negative implicit attitudes toward racial minority students than toward racial majority students (Van den Bergh, Denessen, Hornstra, Voeten, & Holland, 2010). In order to derive hypotheses regarding the nature of implicit attitudes toward racial minority students, we drew on the MODE model, suggesting that implicit attitudes affect judgments. Considering the fact that preservice and inservice teachers judged racial minority students less favorably than racial majority students even when the students showed equal academic achievement (Glock, Krolak-Schwerdt, et al., 2013; Glock & Krolak-Schwerdt, 2013; Parks & Kennedy, 2007), and the relationship between implicit attitudes and judgment specified in the MODE-model, we expected implicit attitudes toward racial minority students to be more negative than implicit attitudes toward racial majority students. This effect should be shown by all implicit attitudes measures. First, we used the affect misattribution procedure AMP (Payne, Cheng, Govorun, & Stewart, 2005; Payne, McClernon, & Dobbins, 2007). Unlike other implicit attitude measures such as the Implicit Association Test IAT (Greenwald, McGhee, & Schwartz, 1998), this task does not rely on reaction times but rather on ratings of stimuli as pleasant or unpleasant (Payne et al., 2007). It has been shown to be unaffected by social pressure in measuring implicit racial attitudes (Payne, Burkley, & Stokes, 2008; Payne et al., 2005). This method is based on the assumption that the attitude object activates a corresponding evaluation, which subsequently results in a judgment about a Chinese pictograph that reflects this evaluation. If the attitude object is positively evaluated, subsequently presented Chinese pictographs will be evaluated as more pleasant than when the attitude object elicits a negative evaluation. Second, we used the IAT, which is currently one of the most prominent measures (Schnabel, Asendorpf, & Greenwald, 2008).
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
The IAT assesses automatic associations and is based on the assumption that people can categorize two concepts more easily if there is an association between the concepts than if there is no association between the concepts. If the attitude object is positively evaluated, people are faster when pairing the attitude object with positive words than when pairing the attitude object with negative words. Third, we employed the affective priming task (Fazio et al., 1995; Fazio, Sanbonmatsu, Powell, & Kardes, 1986). This method is based on the assumption that the attitude object automatically activates the corresponding evaluation, which subsequently facilitates responses to words with valences reflecting the evaluation. Thus, if the attitude object is negatively evaluated, people respond faster to negative than to positive words. We used the same materials as attitude objects in each implicit measurement, and each participant underwent each implicit measurement. This allowed us to compare implicit attitudes toward racial minority students across different measures. Method Participants Sixty-five German preservice teachers (49 female) from the Bergische Universita¨t Wuppertal and the Goethe University Frankfurt participated in the study. They were trained to teach in different German school tracks: Primary school (n = 13), the lower, middle, and highest track of secondary school (n = 49), and the vocational track (n = 3). They were on average 25.23 years old (SD = 3.43) and had on average 5.17 (SD = 8.42) months teaching experience. They participated in partial fulfilment of course requirements. Materials We used the same pictures of male students from racial majorities and racial minorities and the same positive and negative adjectives as used by Glock, Kneer, et al. (2013), integrating these materials into the affective priming task and into the IAT. The pictures displayed 11-year-old smiling male children in halfportrait length. All children were shown in front of the same background. Racial minority pictures displayed ‘‘foreign’’ looking children, since the look of a person helps people to immediately determine ethnicity (Rakic´, Steffens, & Mummendey, 2011). These children had dark hair, dark eyes, and a darker skin than racial majority children. In contrast, the children displayed in the racial majority pictures had no such ‘‘foreign’’ attributes. The pictures had been previously shown to be perceived as ‘‘foreign’’ by German preservice teachers (Glock, Kneer, et al., 2013). The Chinese pictographs implemented in the AMP measure were downloaded from the Inquisit Millisecond Task Library (Millisecond Software). The primes in the AMP measure were the same pictures of male students used in the affective priming task and the IAT. We compiled a demographic questionnaire assessing participants’ age, gender, and teaching experience. Procedure Participants gave informed consent and were seated in front of the computer screen. The three different implicit attitudes measures appeared at random. In the AMP, each trial consisted of a picture followed by a Chinese pictograph. Participants’ task was to rate the pleasantness of the Chinese pictograph. Instructions appeared on the computer screen asking participants to evaluate the Chinese pictographs as pleasant by pressing the ‘‘E’’ key or as unpleasant by pressing the ‘‘I’’ key. Before the beginning
57
of the experimental procedure, participants were informed that the aim of this study was to assess their intuitive judgments of the pictographs and that these pictographs followed a picture. They were warned that the pictures might influence their evaluations but that they should avoid such an influence (Payne et al., 2007). Two different trial types were implemented in the task. One trial type consisted of a picture of a student from a racial minority, which was presented for 75 ms and followed by a Chinese pictograph, which was masked after 200 ms. A second trial type consisted of a picture of a student from a racial majority followed by a Chinese pictograph. The participants’ task was always to rate the pleasantness of the Chinese pictograph, independent of the trial type. Participants performed 10 practice trials before the 52 experimental trials were run in a random sequence. In the first phase of the IAT, participants were asked to categorize the pictures of the students into the categories indicating ‘‘from racial majority’’ by pressing the ‘‘I’’ key and ‘‘from racial minority’’ by pressing the ‘‘E’’ key of the keyboard. In the second phase, participants were presented with positive and negative adjectives and given the task of indicating the word valences by pressing the ‘‘I’’ key for positive and the ‘‘E’’ key for negative words. In the third phase, the two tasks were combined. Participants were asked to press the ‘‘I’’ key in order to indicate ‘‘from racial majority’’ and positive words and the ‘‘E’’ key in order to indicate ‘‘from racial minority’’ and negative words. Participants underwent 40 practice trials in order to become familiar with this combination before the 80 experimental trials were run. In the fourth phase, the combination of the keys was switched. Participants were now asked to press the ‘‘I’’ key for negative and the ‘‘E’’ key for positive words. In the fifth phase, this new assignment of the keys was combined with the pictures. Now, participants were asked to respond with the ‘‘I’’ to negative words and pictures from racial majorities and with the ‘‘E’’ key to positive words and pictures from racial minorities. After 40 practice trials, participants again underwent 80 experimental trials. Half of the participants were presented with the trials combining positive words with racial majority and negative words with racial minority pictures first. The other half of the participants received the trials combining positive words with racial minority and negative words with racial majority picture first. In the affective priming task, participants were informed that they would be presented with pictures followed by words and that their task was to indicate the valence of the words by pressing the ‘‘I’’ key to indicate positive and the ‘‘E’’ key to indicate negative word valences. Four different trials were implemented in this task. Pictures of racial minority students were followed either by positive or negative words, as were racial majority pictures. Participants underwent 10 practice trials, followed by 80 experimental trials. After completing the computerized tasks, participants filled in the demographic questionnaire and were thanked and debriefed. Results Affect misattribution procedure The sum of participants’ pleasant responses was calculated for racial majority and for racial minority pictures. We then divided the sum by the number of possible responses. The resulting value reflected the proportion of pleasant responses, with higher values indicating more positive implicit attitudes (Payne et al., 2007). The scores were submitted to a dependent t-test, t(64) = 4.40, d = 0.55, p < .001, which showed that participants more frequently evaluated the Chinese pictographs as pleasant after racial majority (M = 0.56, SD = 0.22) than after racial minority pictures (M = 0.41, SD = 0.20). Since we had no neutral primes included in the study,
58
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
we compared the frequency of pleasant responses to chance responding (50% of the trials). These single-sample t-tests revealed that participants’ frequency of pleasant responses after racial majority pictures was higher than chance, t(64) = 2.37, d = 0.27, p < .05, whilst the frequency of pleasant responses after minority pictures was lower than chance, t(64) = 3.57, d = 0.45, p < .001. To estimate the internal consistency of the AMP, we randomly selected 20 trials following racial majority pictures and 20 trials following racial minority pictures. We matched majority/minority trials using their order of selection and calculated difference scores for each trial pair. Cronbach’s a computed with the resulting 20 scores was .77. Implicit association test Following the suggestions of Greenwald et al. (1998), we replaced all latencies under 300 ms with 300 ms and above 3000 ms with 3000 ms and included all error trials. This procedure corrected the data for inattention and anticipation (Greenwald et al., 1998). Following the suggestions of Greenwald, Nosek, and Banaji (2003), the practice trials were included in calculating the D measure. First, we calculated the response latencies for trials combining racial majority pictures with positive words and racial minority pictures with negative words, and for trials combining racial majority pictures with negative words and racial minority pictures with positive words. Second, we calculated the D measure as the difference between the response latencies of the two different trial combinations divided by the standard deviation computed from the scores of the two combinations, D measure = 0.93, SD = 0.96. The latencies were submitted to a dependent t-test, t(64) = 7.79, d = 0.97, p < 001. Participants responded more slowly in trials combining racial minority pictures with positive words (M = 1039.99, SD = 221.35) than in trials combining racial minority pictures with negative words (M = 862.26, SD = 204.03). To estimate the internal consistency of the IAT, we randomly selected 20 trials combining racial majority pictures with positive words and 20 trials combining racial minority pictures with positive words. Using their selection order, we matched the two different trial types and calculated the IAT-scores for each of the 20 trial pairs. Cronbach’s a was .59. Affective priming task Response latencies were screened for erroneous classifications and latencies under 250 ms and above 1500 ms (Hermans, Baeyens, Lamote, Spruyt, & Eelen, 2005). These responses were excluded from further analysis (10% of all trials) in order to ensure that only correct and automatic responses were included in the data set. Response latencies were submitted to a 2 2 repeated measures ANOVA with picture (racial majority vs. racial minority) and word (positive vs. negative) as factors. ANOVA revealed a main effect of word, F(1, 64) = 16.03, p < .001, h2p ¼ 0:20, indicating that participants responded faster to positive (M = 716.63, SD = 118.61) than to negative words (M = 747.36, SD = 107.22). ANOVA also yielded a significant interaction, F(1, 64) = 13.08, p < .001, h2p ¼ 0:17. Participants responded more quickly to positive words (M = 703.39, SD = 121.39) than to negative words (M = 757.53, SD = 118.12) following racial majority pictures, t(64) = 5.57, d = 0.69, p < .001. Following racial minority pictures, there was no difference between positive (M = 729.86, SD = 125.40) and negative words (M = 737.19, SD = 106.90), t(64) = 0.71, d = 0.09, p = .48. Participants responded more quickly to positive words following racial majority than racial minority pictures, t(64) = 3.13, d = 0.39, p < .01, and more quickly to negative following racial minority than racial majority pictures, t(64) = 2.37, d = 0.29,
Table 1 Effect sizes d of each implicit measure. Implicit measure
IAT
AMP
APT (interaction)
Effect size d
0.93
0.55
0.91
p < .05. To determine the internal consistency of the affective priming task, we randomly ordered the four types of trials to create clusters of four response latencies (one for each trial type). Within each cluster, we first calculated two difference scores by subtracting the response latencies for positive words from the latencies for negative words following either racial majority or racial minority pictures. Then we subtracted the difference score for racial majority pictures from the difference score for racial minority pictures. The resulting 20 scores were used to compute Cronbach’s a = .42. Table 1 presents the effect size d of each measure. In a last step, we calculated the correlations between the three implicit measures. Thus, we transformed the values of each implicit measure into one single score. We calculated (1) the AMP score by subtracting the responses for racial majority from the responses for racial minority pictures. Positive values on this score reflected positive attitudes toward racial minorities (range: .88–.12). We calculated (2) the IAT score by subtracting the incompatible trials from the compatible trials. Positive values on this IAT score reflected positive attitudes toward racial minorities (range: 599.82– 313.80). We calculated the priming difference score using the same procedure as described for the calculation of the internal consistency of the affective priming task, except that we now relied on the means of each trial combination. Positive values on this priming difference score indicated positive attitudes toward racial minorities (range: 169.90–346.32). Intercorrelations are displayed in Table 2. The different implicit attitudes measures showed no significant correlations.
Discussion The AMP as well as the IAT revealed negative implicit attitudes toward racial minority. In the AMP, this was indicated by a higher proportion of pleasant responses as well as by the finding that the frequency of pleasant responses after racial minority pictures was lower than by chance. In the IAT, this can be concluded from the faster responses in compatible trials combining racial majority students with positive and racial minority students with negative words than in incompatible trials. However, a closer inspection of the results of the affective priming task revealed that participants showed positivity toward racial majority students rather than negativity toward racial minority students. This was indicated by faster responses to positive than to negative words following racial majority students, which is also in line with previous research (Glock, Kneer, et al., 2013). The affective priming task allows us to draw conclusions about the implicit attitudes toward racial minority as well as toward racial majority students (De Houwer, Teige-Mocigemba, Spruyt, & Moors, 2009), while the IAT only provides a single comparative measure indicating the difference in attitudes toward these two groups. Moreover, although we had no neutral primes included in the AMP measure, we might Table 2 Intercorrelations and p-values in parentheses between the three implicit attitudes measures.
AMP IAT Note: *p < .05.
IAT
Affective priming
.07 (.57)
.06 (.64) .09 (.47)
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
nevertheless be able to disentangle implicit attitudes toward racial minority students from those toward racial majority students (Dunham & Emory, 2014). Thus, from the AMP measure and the comparison of the frequency of pleasant responses to chance, we can cautiously conclude that the lower proportion of pleasant responses is due to negative implicit attitudes toward racial minority students as well as to positive attitudes toward racial majority students. These results partially support and partially stand in contrast to the findings of the affective priming measure. Hence, future research is needed to investigate why these two implicit measures revealed different implicit attitudes toward racial minority students. Including neutral primes in both measures in future research might help us gain a deeper understanding of the discrepancies in our results. One difference between the AMP and the affective priming task is that the former relies on ratings while the latter relies on response latencies (Payne et al., 2007). Compared to the affective priming task, the AMP is superior in reliability and in predictive validity (Payne, 2009). Nevertheless, on order to disentangle attitudes toward racial minority students from those toward majority students, future research should employ additional implicit reliable measures such as the Single-Category IAT (Karpinski & Steinman, 2006) or the Extrinsic Affective Simon Task EAST (De Houwer, 2003), which also allows to disentangle attitudes toward racial minority from those toward majority. This could shed light on the question of whether implicit positivity toward racial majority, implicit negativity toward racial minority students, or both caused the effects. The three implicit attitudes measures showed no substantial correlations, which is not entirely surprising, given that correlations between the IAT and the affective priming task are generally low (Olson & Fazio, 2003; Payne, Govorun, et al., 2008). The IAT measures attitudes toward categories while the affective priming task measures attitudes toward specific exemplars which are then averaged into the categories (Olson & Fazio, 2003). Thus, the correlation between the two implicit measures increases when the affective priming task also requires a categorization (Olson & Fazio, 2003). Although the AMP as well as the affective priming task are priming measures, the AMP relies on pleasant and unpleasant responses, while the affective priming tasks relies on response latencies, which might partially explain the low correlations. A study employing the same three implicit measures of attitudes also reported no significant correlations between the AMP, the IAT, and the affective priming task (Payne, Govorun, et al., 2008). Interestingly, this study also implemented consistent stimulus materials into all three measures, thus, the low correlations did not stem from variations in the material used. The low correlations we found between different implicit measures are in line with previous research (e.g., Olson & Fazio, 2003; Sherman et al., 2003) and might – to some extent – reflect measurement error (Cunningham, Preacher, & Banaji, 2001; De Houwer et al., 2009) as well as the different structure underlying each implicit measure (De Houwer, 2009). The reliability values of the three measures support this assumption. Although the AMP showed rather good reliability, the IAT showed moderate and the affective priming task low reliability values. Thus, the low correlations might indeed partially stem from measurement error, however, it also might be that – besides the different structure of the implicit measures – they tap different facets of implicit attitudes reflecting the complexity of implicit attitudes (Bosson, Swann, & Pennebaker, 2000). Nonetheless, our results have implications for teachers’ and preservice teachers’ judgments. Although preservice teachers implicit attitudes did not differ from the general population (Nosek, Banaji, & Greenwald, 2002), their attitudes are of particular interest. These educators will assign grades and decide about the school career of students. Thus, preservice teachers’ attitudes could play a pivotal role for the educational career of the students.
59
Given the negative racial bias found in the IAT and the AMP, all three implicit measures imply more favorable preservice teachers’ attitudes, judgments, and behaviors toward racial majority students. This implicit racial bias might contribute to disadvantages experienced by racial minority students. Since teachers often compare the academic achievement of a student with the achievement of his or her peers in the class (Su¨dkamp et al., 2012), the implicit racial bias may play a pivotal role when judging a racial minority student in comparison to a racial majority student. This assumption is in line with previous research showing that, compared to a racial majority student displaying the same academic profile, racial minority students are judged less favorably (Glock, Krolak-Schwerdt, et al., 2013; Glock & Krolak-Schwerdt, 2013). Notwithstanding the implicit racial bias we found in this study, implicit attitudes have been shown to be malleable (Blair, 2002; Dasgupta & Greenwald, 2001) with the attitude object stored in a multifaceted manner (Gawronski & Bodenhausen, 2006). The context in which implicit attitudes are activated may play a large role in determining which automatic evaluation is activated (Fazio, 2007). In our study, we employed no specific cue to activate a particular evaluation, meaning that the resulting measures can be seen as ‘‘baseline’’ attitudes. Previous research has shown that confronting participants with positive Black exemplars resulted in more positive implicit attitudes toward Black individuals than confronting participants with a negative Black exemplar (Dasgupta & Greenwald, 2001). In this vein, implicit racial attitudes can be changed through diversity education (Rudman, Ashmore, & Gary, 2001). Therefore, it might be valuable for future research to investigate whether confronting preservice teachers with an expectation-confirming racial minority student can turn those ambivalent implicit attitudes into more negative ones and whether more positive implicit attitudes can be induced by presenting participants with an expectation-disconfirming racial minority student. Previous research has shown that explicitly assessed stereotypical beliefs are more positive after preservice teachers have been confronted with an expectation-disconfirming racial minority student (Glock & Krolak-Schwerdt, 2013). Moreover, it has also been shown that racial minority students are only judged less favorably when they are prototypical members of this social group (Glock & Krolak-Schwerdt, 2013; Parks & Kennedy, 2007). Therefore, when preservice teachers enter the classroom and are confronted with an expectation-confirming racial minority student, activated negative implicit attitudes might affect their judgments, possibly leading to less favorable judgments even though this students’ academic performance is equal to that of racial majority students. Our study has some limiting aspects that should be kept in mind. As the focus of our study was to employ different implicit measures of attitudes in order to investigate the nature of implicit attitudes toward racial minority students, we did not assess explicit attitudes. Since explicit attitudes measures rely on selfreports, they may – to a substantial extent – reflect social norms and social desirability rather than explicit attitudes. However, future research should also employ an explicit attitudes measure, such as the Teacher Cultural Beliefs Scale (Hachfeld et al., 2011). This would shed further light on possible differences between implicit and explicit attitudes, thus providing a deeper understanding of teachers’ attitudes. We did not assess actual behavior, judgments, or even behavioral intentions. Thus, future research should relate implicit attitudes to actual behavior, judgments, or behavioral intentions in order to investigate whether implicit attitudes contribute to judgment biases. Since implicit attitudes are automatically activated by the mere presence of the attitude object (Fazio, 2001), it seems plausible that preservice teachers’ behavior might be affected by implicit attitudes. Particularly, given that people are
60
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
often not aware of this influence (Asendorpf et al., 2002) and that implicit attitudes exhibit a greater influence when cognitive resources are limited (Fazio, 1990; Olson & Fazio, 2009), classroom behavior is likely to be guided by implicit attitudes. No experienced teachers were included in our sample. Thus, our results are limited to preservice teachers and we cannot draw conclusions about the nature of experienced teachers’ implicit attitudes toward racial minority students. One study exploring implicit attitudes toward racial minority students found slightly negative implicit attitudes, which were related to the academic achievement of the students assessed via standardized tests (Van den Bergh et al., 2010). Racial minority students instructed by teachers with more negative implicit attitudes achieved lower test scores than racial minority students instructed by teachers with more positive implicit attitudes (Van den Bergh et al., 2010). This research emphasizes the pivotal role of implicit attitudes in the classroom routine – and consequently the vital role of research on experienced teachers’ implicit attitudes toward racial minority students. Notwithstanding the limitations of our study, the employment of three different implicit attitudes measures provided a consistent pattern of implicit attitudes that revealed more positive attitudes toward racial majority than toward racial minority students. Keeping in mind that preservice teachers are future teachers, our study provides valuable insights into the nature of their implicit attitudes, which could be used to implement specific training courses early during their academic studies in order to overcome implicit positivity toward racial majority students and help ensure the fair treatment of all students.
Author note We would like to thank Carrie Kovacs for proof reading the manuscript.
References Ansalone, G. (2001). Schooling, tracking, and inequality. Journal of Children and Poverty, 7, 33–47. http://dx.doi.org/10.1080/10796120120038028 Ansalone, G., & Biafora, F. (2004). Elementary school teachers’ perceptions and attitudes to the educational structure of tracking. Education, 125, 249–260. Asendorpf, J. B., Banse, R., & Mu¨cke, D. (2002). Double dissociation between implicit and explicit personality self-concept: The case of shy behavior. Journal of Personality and Social Psychology, 83, 380–393. http://dx.doi.org/10.1037//0022-3514.83.2.380 Blair, I. (2002). The malleability of automatic stereotypes and prejudice. Personality and Social Psychology Review, 6, 242–261. http://dx.doi.org/10.1207/S15327957 PSPR0603_8 Bosson, J. K., Swann, W. B., & Pennebaker, J. W. (2000). Stalking the perfect measure of implicit self-esteem: The blind men and the elephant revisited? Journal of Personality and Social Psychology, 79, 631–643. Caro, D. H., Lenkeit, J., Lehmann, R., & Schwippert, K. (2009). The role of academic achievement growth in school track recommendations. Studies in Educational Evaluation, 35, 183–192. http://dx.doi.org/10.1016/j.stueduc.2009.12.002 Coneus, K., Gernandt, J., & Saam, M. (2009). Noncognitive skills, school achievements and educational dropout. ZEW Discussion Papers, No. 09-019. Cunningham, W. A., Preacher, K. J., & Banaji, M. R. (2001). Implicit attitude measures: Consistency, stability, and convergent validity. Psychological Science, 12, 163–170. Dasgupta, N., & Greenwald, A. G. (2001). On the malleability of automatic attitudes: Combating automatic prejudice with images of admired and disliked individuals. Journal of Personality and Social Psychology, 81, 800–814. http://dx.doi.org/10.1037/ 0022-3514.81.5.800 De Houwer, J. (2003). The extrinsic affective Simon task. Experimental Psychology, 50, 77–85. http://dx.doi.org/10.1027//1618-3169.50.2.77 De Houwer, J. (2006). What are implicit measures and why are we using them? In R. W. Wiers & A. W. Stacy (Eds.), Handbook of implicit cognition and addiction (pp. 11–28). Thousand Oaks, CA: Sage Publisher. De Houwer, J. (2009). Comparing measures of attitudes at the functional and procedural level: Analysis and implications. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 361–390). New York, NY: Psychology Press. De Houwer, J., Geldof, T., De Bruycker, E., & De Bruycher, E. (2005). The implicit association test as a general measure of similarity. Canadian Journal of Experimental Psychology, 59, 228–239. http://dx.doi.org/10.1037/h0087478
De Houwer, J., Teige-Mocigemba, S., Spruyt, A., & Moors, A. (2009). Implicit measures: A normative analysis and review. Psychological Bulletin, 135, 347–368. http:// dx.doi.org/10.1037/a0014211 Dee, T. S. (2005). A teacher like me: Does race, ethnicity, or gender matter? The American Economic Review, 95, 158–165. http://dx.doi.org/10.1257/00028280 5774670446 Destatis (2012). Ausla¨ndische Bevo¨lkerung [foreign population]. Available from: https://www.destatis.de/DE/ZahlenFakten/GesellschaftStaat/Bevoelkerung/ MigrationIntegration/AuslaendischeBevolkerung/Tabellen/Staatsange hoerigkeitJahre.html Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18. http:// dx.doi.org/10.1037//0022-3514.56.1.5 Dovidio, J. F., Brigham, J. C., Johnson, B. S., & Gaertner, S. L. (1996). Stereotyping, prejudice, and discrimination: Another look. In C. N. Macrae, M. Hewstone, & C. Stangor (Eds.), Stereotypes and stereotyping (pp. 276–319). New York: Guilford Press. Dovidio, J. F., & Fazio, R. H. (1992). New technologies for the direct and indirect assessment of attitudes. In J. M. Tanur (Ed.), Questions about questions: Inquiries into the cognitive bases of surveys (pp. 204–237). New York, NY: Russel Sage Foundation. Dunham, Y., & Emory, J. (2014). Of affect and ambiguity: The emergence of preference for arbitrary ingroups. Journal of Social Issues, 70, 81–98. http://dx.doi.org/10.1111/ josi.12048 Euwals, R., Dagevos, J., Gijsberts, M., & Roodenburg, H. (2007). Immigration, integration, and the labour market: Turkish immigrants in Germany and the Netherlands. Discussion Paper no. 2677. Bonn: IZA. Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (23, pp. 75–109). New York, NY: Academic Press. Fazio, R. H. (2001). On the automatic activation of associated evaluations: An overview. Cognition and Emotion, 15, 115–141. http://dx.doi.org/10.1080/026999300 4200024 Fazio, R. H. (2007). Attitudes as object-evaluation associations of varying strength. Social Cognition, 25, 603–637. http://dx.doi.org/10.1521/soco.2007.25.5.603 Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology, 69, 1013–1027. http://dx.doi.org/ 10.1037/0022-3514.69.6.1013 Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229–238. http://dx.doi.org/10.1037//0022-3514.50.2.229 Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change. Psychological Bulletin, 132, 692–731. http://dx.doi.org/10.1037/0033-2909.132.5.692 Gawronski, B., Strack, F., & Bodenhausen, G. V. (2009). Attitudes and cognitive consistency. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 85–117). New York, NY: Psychology Press. Glock, S., Kneer, J., & Kovacs, C. (2013). Preservice teachers’ implicit attitudes toward students with and without immigration background: A pilot study. Studies in Educational Evaluation, 39, 204–210. http://dx.doi.org/10.1016/j.stueduc.2013. 09.003 Glock, S., & Kovacs, C. (2013). Educational psychology: Using insights from implicit attitude measures. Educational Psychology Review, 25, 503–522. http://dx.doi.org/ 10.1007/s10648-013-9241-3 Glock, S., & Krolak-Schwerdt, S. (2013). Does nationality matter? The impact of stereotypical expectations on student teachers’ judgments. Social Psychology of Education, 16, 111–127. http://dx.doi.org/10.1007/s11218-012-9197-z Glock, S., Krolak-Schwerdt, S., Klapproth, F., & Bo¨hmer, M. (2013). Beyond judgment bias: How students’ ethnicity and academic profile consistency influence teachers’ tracking judgments. Social Psychology of Education, 16, 555–573. http://dx.doi.org/ 10.1007/s11218-013-9227-5 Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74, 1464–1480. http://dx.doi.org/10.1037/0022-3514.74.6. 1464 Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the implicit association test I: An improved scoring algorithm. Journal of Personality and Social Psychology, 85, 197–216. http://dx.doi.org/10.1037/0022-3514.85.2. 197 Hachfeld, A., Hahn, A., Schroeder, S., Anders, Y., Stanat, P., & Kunter, M. (2011). Assessing teachers’ multicultural and egalitarian beliefs: The teacher cultural beliefs scale. Teaching and Teacher Education, 27, 986–996. http://dx.doi.org/10. 1016/j.tate.2011.04.006 Haycock, K. (2001). Closing the achievement gap. Educational Leadership, 58, 6–11. Hermans, D., Baeyens, F., Lamote, S., Spruyt, A., & Eelen, P. (2005). Affective priming as an indirect measure of food preferences acquired through odor conditioning. Experimental Psychology, 52, 180–186. http://dx.doi.org/10.1027/1618-3169.52.3. 180 Hofmann, W., Gschwendner, T., Nosek, B. A., & Schmitt, M. (2005). What moderates implicit—Explicit consistency? European Review of Social Psychology, 16, 335–390. http://dx.doi.org/10.1080/10463280500443228 Jussim, L., & Harber, K. D. (2005). Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9, 131–155. http://dx.doi.org/10.1207/s15327957 pspr0902_3
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61 Karpinski, A., & Hilton, J. L. (2001). Attitudes and the implicit association test. Journal of Personality and Social Psychology, 81, 774–788. http://dx.doi.org/10.1037//00223514.81.5.774 Karpinski, A., & Steinman, R. B. (2006). The single category implicit association test as a measure of implicit social cognition. Journal of Personality and Social Psychology, 91, 16–32. http://dx.doi.org/10.1037/0022-3514.91.1.16 Kristen, C., & Granato, N. (2007). The educational attainment of the second generation in Germany: Social origins and ethnic inequality. Ethnicities, 7, 343–366. http:// dx.doi.org/10.1177/1468796807080233 Lee, J. (2002). Racial and ethnic achievement gap trends: Reversing the progress toward equity. Educational Researcher, 31, 3–12. http://dx.doi.org/10.3102/ 0013189X031001003 Lucas, S. R. (2001). Effectively maintained inequality: Education transitions, track mobility, and social background effects. American Journal of Sociology, 106, 1642–1690. http://dx.doi.org/10.1086/321300 McCombs, R. C., & Gay, J. (1988). Effects of race, class, and IQ information on judgments of parochial grade school teachers. The Journal of Social Psychology, 128, 647–652. http://dx.doi.org/10.1080/00224545.1988.9922918 Millisecond Software. (n.d.). Inquisit task library. Available from http://www. millisecond.com/download/library/AMP/. Moors, A., & De Houwer, J. (2006). Automaticity: A theoretical and conceptual analysis. Psychological Bulletin, 132, 297–326. http://dx.doi.org/10.1037/0033-2909.132.2. 297 Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). Harvesting implicit group attitudes and beliefs from a demonstration web site. Group Dynamics: Theory, Research, and Practice, 6, 101–115. http://dx.doi.org/10.1037//1089-2699.6.1.101 Oakes, J. (2005). Keeping track: How schools structure inequality (2nd ed.). New Haven, CT: Yale University Press. Olson, M. A., & Fazio, R. H. (2003). Relations between implicit measures of prejudice: What are we measuring? Psychological Science, 14, 636–639. http://dx.doi.org/ 10.1046/j.0956-7976.2003.psci_1477.x Olson, M. A., & Fazio, R. H. (2009). Implicit and explicit measures of attitudes: The perspective of the MODE model. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 19–63). New York, NY: Psychology Press. Parks, F. R., & Kennedy, J. H. (2007). The impact of race, physical attractiveness, and gender on education majors’ and teachers’ perceptions of student competence. Journal of Black Studies, 37, 936–943. http://dx.doi.org/10.1177/0021934705285955 Payne, B. K. (2009). Attitude misattribution: Implications for attitude measurement and the implicit-explicit relationship. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures. Psychology Press: New York, NY. Payne, B. K., Burkley, M., & Stokes, M. B. (2008). Why do implicit and explicit attitude tests diverge? The role of structural fit. Journal of Personality and Social Psychology, 94, 16–31. http://dx.doi.org/10.1037/0022-3514.94.1.16 Payne, B. K., Cheng, S. M., Govorun, O., & Stewart, B. D. (2005). An inkblot for attitudes: Affect misattribution as implicit measurement. Journal of Personality and Social Psychology, 89, 277–293. http://dx.doi.org/10.1037/0022-3514.89.3.277 Payne, B. K., Govorun, O., & Arbuckle, N. L. (2008). Automatic attitudes and alcohol: Does implicit liking predict drinking? Cognition & Emotion, 22, 238–271. http:// dx.doi.org/10.1080/02699930701357394 Payne, B. K., McClernon, F. J., & Dobbins, I. G. (2007). Automatic affective responses to smoking cues. Experimental and Clinical Psychopharmacology, 15, 400–409. http:// dx.doi.org/10.1037/1064-1297.15.4.400 Rakic´, T., Steffens, M. C., & Mummendey, A. (2011). Blinded by the accent! The minor role of looks in ethnic categorization Journal of Personality and Social Psychology, 100, 16–29. http://dx.doi.org/10.1037/a0021522
61
Robinson, M. D., Meier, B. P., Zetocha, K. J., & McCaul, K. D. (2005). Smoking and the implicit association test: When the contrast category determines the theoretical conclusions. Basic and Applied Social Psychology, 27, 201–212. http://dx.doi.org/ 10.1207/s15324834basp2703_2 Rudman, L. A., Ashmore, R. D., & Gary, M. L. (2001). Unlearning, automatic biases: The malleability of implicit prejudice and stereotypes. Journal of Personality and Social Psychology, 81, 856–868. http://dx.doi.org/10.1037//0022-3514.81.5.856 Sanbonmatsu, D. M., & Fazio, R. H. (1990). The role of attitudes in memory-based decision making. Journal of Personality and Social Psychology, 59, 614–622. http:// dx.doi.org/10.1037//0022-3514.59.4.614 Santavirta, N., Solovieva, S., & Theorell, T. (2007). The association between job strain and emotional exhaustion in a cohort of 1.028 Finnish teachers. British Journal of Educational Psychology, 77, 213–228. http://dx.doi.org/10.1348/000709905X92045 Schnabel, K., Asendorpf, J. B., & Greenwald, A. G. (2008). Assessment of individual differences in implicit cognition. European Journal of Psychological Assessment, 24, 210–217. http://dx.doi.org/10.1027/1015-5759.24.4.210 Sherman, S. J., Rose, J. S., Koch, K., Presson, C. C., & Chassin, L. (2003). Implicit and explicit attitudes toward cigarette smoking: The effects of context and motivation. Journal of Social and Clinical Psychology, 22, 13–39. http://dx.doi.org/10.1521/ jscp.22.1.13.22766 Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220–247. http://dx.doi.org/ 10.1207/s15327957pspr0803_1 Su¨dkamp, A., Kaiser, J., & Mo¨ller, J. (2012). Accuracy of teachers’ judgments of students’ academic achievement: A meta-analysis. Journal of Educational Psychology, 104, 743–762. http://dx.doi.org/10.1037/a0027627 Tenenbaum, H. R., & Ruck, M. D. (2007). Are teachers’ expectations different for racial minority than for European American students? A meta-analysis. Journal of Educational Psychology, 99, 253–273. http://dx.doi.org/10.1037/0022-0663.99.2.253 Van den Bergh, L., Denessen, E., Hornstra, L., Voeten, M., & Holland, R. W. (2010). The implicit prejudiced attitudes of teachers: Relations to teacher expectations and the ethnic achievement gap. American Educational Research Journal, 47, 497–527. http://dx.doi.org/10.3102/0002831209353594 Van Dick, R., & Wagner, U. (2001). Stress and strain in teaching: A structural equation approach. British Journal of Educational Psychology, 71, 243–259. http://dx.doi.org/ 10.1348/000709901158505 Wilson, T. D., Lindsey, S., & Schooler, T. Y. (2000). A model of dual attitudes. Psychological Review, 107, 101–126. http://dx.doi.org/10.1037//0033-295X.107.1.101
Sabine Glock is an educational psychologist holding a PhD in psychology from the University of Saarland. Since 2014, she is a member of the Institute for Educational Research in the School of Education at the Bergische Universita¨t Wuppertal, Germany. She is concerned with educational decision making and factors influencing educational decisions. Her main research interest is implicit cognition, in particular implicit attitudes toward students with ethnic/racial minority background and how they contribute to teacher judgment
Julia Karbach is a cognitive developmental psychologist holding a PhD in psychology from the University of Saarland. She currently serves as professor at Goethe University Frankfurt, Germany. Her main research interests include cognitive development and plasticity across the lifespan as well as teacher education. Her work has been published in leading peer-reviewed international journals.
Contents lists available at ScienceDirect
Studies in Educational Evaluation journal homepage: www.elsevier.com/stueduc
Preservice teachers’ implicit attitudes toward racial minority students: Evidence from three implicit measures Sabine Glock a,*, Julia Karbach b a b
Bergische Universita¨t Wuppertal, Germany Goethe University Frankfurt, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 June 2014 Received in revised form 16 March 2015 Accepted 17 March 2015 Available online 11 April 2015
Implicit attitudes can be activated by the mere presence of the attitude object. They are assumed to guide behavior in demanding situations, including the educational context. Implicit attitudes toward racial minority students could be important in contributing to the disadvantages those students experience in school. This study employed three different measures to investigate implicit attitudes toward racial minority students among preservice teachers. The IAT and the AMP showed more negative implicit attitudes toward racial minority than toward racial majority students; the affective priming task revealed that implicit attitudes toward racial majority students were positive, while those toward racial minority students were neutral. Results are discussed in their implications for preservice teachers’ judgments and behaviors. ß 2015 Elsevier Ltd. All rights reserved.
Keywords: Racial minority students Implicit attitudes Implicit measures Preservice teachers
Introduction Educational research has consistently shown that racial minority students experience disadvantages in educational systems. Compared to racial majority students, they perform poorly in school (e.g., Dee, 2005; Haycock, 2001; Lee, 2002) and thus are overrepresented in lower level and vocational courses (Ansalone, 2001; Lucas, 2001; Oakes, 2005), in lower school tracks (Caro, Lenkeit, Lehmann, & Schwippert, 2009). They are also recommended less frequently for the higher level tracks, even when academic performance is controlled for (Glock, KrolakSchwerdt, Klapproth, & Bo¨hmer, 2013). In Germany, racial minority students mainly stem from Turkey (Destatis, 2012). Students with Turkish roots are overrepresented on the lowest school track (Caro et al., 2009) and underrepresented on the highest track (Kristen & Granato, 2007). They more frequently fail to complete school (Coneus, Gernandt, & Saam, 2009) and consequently, they have jobs with lower prestige and lower employment rates than their German peers (Euwals, Dagevos, Gijsberts, & Roodenburg, 2007). In the German school system, attending the highest school track and leaving this track
* Corresponding author at: Bergische Universita¨t Wuppertal, School of Education, Gaußstraße 20, 42119 Wuppertal, Germany. Tel.: +49 0202 439 3082. E-mail address: [email protected] (S. Glock). http://dx.doi.org/10.1016/j.stueduc.2015.03.006 0191-491X/ß 2015 Elsevier Ltd. All rights reserved.
with a qualification for university entrance is of high importance for the future professional career of students. Considering that teachers are the main decision makers when it comes to grading or tracking (Ansalone & Biafora, 2004), the abovementioned disadvantages might not only stem from racial minority students’ lower academic achievement, but also from biases in teachers’ judgments. In educational research, stereotypes and teacher expectations are discussed as factors influencing teachers’ judgments (e.g., Jussim & Harber, 2005; Su¨dkamp, Kaiser, & Mo¨ller, 2012). To this extent, teachers expect racial minority students to show lower academic achievement than racial majority students (Tenenbaum & Ruck, 2007). Teacher judgments have been shown to be affected by race (McCombs & Gay, 1988; Parks & Kennedy, 2007), indicating a possible negative bias against the racial background of a student; this was true for both experienced and preservice teachers with limited teaching experience. In this vein, negative teacher and preservice teacher biases might reflect prejudice (Devine, 1989), defined as negative attitudes toward the members of a social group (Dovidio, Brigham, Johnson, & Gaertner, 1996). Attitudes reflect the positive or negative evaluation associated with an object or a social group (Fazio, 2007) and attitudes might affect how people are perceived and judged (Olson & Fazio, 2009; Sanbonmatsu & Fazio, 1990). Thus, it seems of high importance to investigate the nature of attitudes toward racial minority students, as those attitudes might be reflected in judgments. Hence, the aim of this study was to
56
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
investigate preservice teachers’ attitudes toward racial minority students.
Theoretical background On a theoretical level, implicit and explicit attitudes are distinct constructs resulting from two different mental processes (Gawronski, Strack, & Bodenhausen, 2009). Explicit attitudes are defined as conscious evaluations of the attitude object which result from reflected and controlled processes (Gawronski & Bodenhausen, 2006). In contrast, implicit attitudes are defined as automatic evaluations of the attitude object which result from automatic processes (Gawronski & Bodenhausen, 2006). Implicit attitudes can be activated by the presence of the attitude object (Fazio, 2001). Thus, the different kinds of attitudes resulting from different mental processes are prevalent in different situations. These processes are specified in dual process models on how attitudes guide behaviors (e.g., Fazio, 1990; Gawronski & Bodenhausen, 2006; Strack & Deutsch, 2004; Wilson, Lindsey, & Schooler, 2000). In this study, we draw on the Motivation and Opportunity as Determinants (MODE) model (Fazio, 1990; Olson & Fazio, 2009). Explicit attitudes are assumed to guide perception, judgments, and behavior in situations where people have enough time, cognitive resources, and motivation to reflect on their attitudes, make conscious judgments, and control their behavior (Fazio, 1990; Olson & Fazio, 2009). In situations with high cognitive demands, implicit attitudes should be prevalent and guide perception, judgments, and behavior. Although these two processes are not mutually exclusive and both explicit as well as implicit attitudes might come into play (Olson & Fazio, 2009), the automatic character of implicit attitudes means that implicit attitudes are dominant in most situations, as they are automatically activated by the mere presence of the attitude object (Fazio, 2001) and people are often not aware of this influence (Asendorpf, Banse, & Mu¨cke, 2002). Implicit and explicit attitudes are measured using quite distinct methods. Explicit attitudes measures rely on the assumption that explicit attitudes are conscious evaluations and that people can report their explicit attitudes when asked to evaluate an object. Thus, explicit attitudes are assessed using questionnaires and rely on self-report measures. However, when it comes to socially sensitive issues, people are often reluctant to report their ‘‘real’’ explicit attitudes (Dovidio & Fazio, 1992; Fazio, Jackson, Dunton, & Williams, 1995), and explicit attitudes measures might reflect social norms (Fazio et al., 1995; Karpinski & Hilton, 2001) and social desirability (De Houwer, 2006) rather than explicit attitudes. Hence, implicit attitudes measures attempt to reduce the social desirability bias (De Houwer, 2006). Since implicit attitudes are defined as being automatic, implicit attitudes measures tap into automaticity (De Houwer, 2006; Moors & De Houwer, 2006). The definition of automatic processes involves unconsciousness, nonintention, unawareness, and efficiency. Therefore, implicit measures usually share at least one of these properties (Hofmann, Gschwendner, Nosek, & Schmitt, 2005). As a whole, implicit attitudes measures seem to be a valuable tool in assessing socially sensitive issues such as attitudes toward racial minority students. Implicit attitudes may also play a pivotal role for preservice teachers’ judgments and behaviors, since working as a teacher is stressful (Van Dick & Wagner, 2001), often requiring action under time pressure (Santavirta, Solovieva, & Theorell, 2007). Under such conditions, implicit attitudes may be particularly influential. Given that preservice teachers’ judgments can have a great impact on students’ future educational and professional careers, preservice teachers’ implicit attitudes toward racial minority students are vital when they enter the classroom.
In the last years, many different implicit measures have been developed (see Glock & Kovacs, 2013, for an overview of most but not all measures). Since previous research on implicit attitudes has shown that the nature of implicit attitudes differs as a function of the stimuli used (Robinson, Meier, Zetocha, & McCaul, 2005) and the methods applied (Sherman, Rose, Koch, Presson, & Chassin, 2003), inconsistent results regarding implicit attitudes toward racial minority students could stem from those measurement methods. Although research employing the same three implicit measures has not found substantial differences in revealed implicit attitudes, it is nevertheless plausible that each measure assesses unique aspects of implicit attitudes (Payne, Govorun, & Arbuckle, 2008). Moreover, each implicit measure requires different categorization tasks (Olson & Fazio, 2003) and participants’ performance might not only reflect automatic attitudes, but also particular features of the stimuli or of the categories employed in the measure (De Houwer, Geldof, De Bruycker, & De Bruycher, 2005; De Houwer, 2003). Hence, employing multiple measures of implicit attitudes is highly recommended in implicit attitudes research. Therefore, in order to rule out that measurement methods affect the results to a greater extent than implicit attitudes do, we employed three different implicit attitudes measures using the same stimulus materials. Research question The aim of this study was the investigation of the nature of implicit attitudes toward racial minority students. Educational research on implicit attitudes toward racial minority students is particularly sparse. One study found ambivalent implicit attitudes toward racial minority students among preservice teachers, while the attitudes toward racial majority students were positive (Glock, Kneer, & Kovacs, 2013). Another study found slightly more negative implicit attitudes toward racial minority students than toward racial majority students (Van den Bergh, Denessen, Hornstra, Voeten, & Holland, 2010). In order to derive hypotheses regarding the nature of implicit attitudes toward racial minority students, we drew on the MODE model, suggesting that implicit attitudes affect judgments. Considering the fact that preservice and inservice teachers judged racial minority students less favorably than racial majority students even when the students showed equal academic achievement (Glock, Krolak-Schwerdt, et al., 2013; Glock & Krolak-Schwerdt, 2013; Parks & Kennedy, 2007), and the relationship between implicit attitudes and judgment specified in the MODE-model, we expected implicit attitudes toward racial minority students to be more negative than implicit attitudes toward racial majority students. This effect should be shown by all implicit attitudes measures. First, we used the affect misattribution procedure AMP (Payne, Cheng, Govorun, & Stewart, 2005; Payne, McClernon, & Dobbins, 2007). Unlike other implicit attitude measures such as the Implicit Association Test IAT (Greenwald, McGhee, & Schwartz, 1998), this task does not rely on reaction times but rather on ratings of stimuli as pleasant or unpleasant (Payne et al., 2007). It has been shown to be unaffected by social pressure in measuring implicit racial attitudes (Payne, Burkley, & Stokes, 2008; Payne et al., 2005). This method is based on the assumption that the attitude object activates a corresponding evaluation, which subsequently results in a judgment about a Chinese pictograph that reflects this evaluation. If the attitude object is positively evaluated, subsequently presented Chinese pictographs will be evaluated as more pleasant than when the attitude object elicits a negative evaluation. Second, we used the IAT, which is currently one of the most prominent measures (Schnabel, Asendorpf, & Greenwald, 2008).
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
The IAT assesses automatic associations and is based on the assumption that people can categorize two concepts more easily if there is an association between the concepts than if there is no association between the concepts. If the attitude object is positively evaluated, people are faster when pairing the attitude object with positive words than when pairing the attitude object with negative words. Third, we employed the affective priming task (Fazio et al., 1995; Fazio, Sanbonmatsu, Powell, & Kardes, 1986). This method is based on the assumption that the attitude object automatically activates the corresponding evaluation, which subsequently facilitates responses to words with valences reflecting the evaluation. Thus, if the attitude object is negatively evaluated, people respond faster to negative than to positive words. We used the same materials as attitude objects in each implicit measurement, and each participant underwent each implicit measurement. This allowed us to compare implicit attitudes toward racial minority students across different measures. Method Participants Sixty-five German preservice teachers (49 female) from the Bergische Universita¨t Wuppertal and the Goethe University Frankfurt participated in the study. They were trained to teach in different German school tracks: Primary school (n = 13), the lower, middle, and highest track of secondary school (n = 49), and the vocational track (n = 3). They were on average 25.23 years old (SD = 3.43) and had on average 5.17 (SD = 8.42) months teaching experience. They participated in partial fulfilment of course requirements. Materials We used the same pictures of male students from racial majorities and racial minorities and the same positive and negative adjectives as used by Glock, Kneer, et al. (2013), integrating these materials into the affective priming task and into the IAT. The pictures displayed 11-year-old smiling male children in halfportrait length. All children were shown in front of the same background. Racial minority pictures displayed ‘‘foreign’’ looking children, since the look of a person helps people to immediately determine ethnicity (Rakic´, Steffens, & Mummendey, 2011). These children had dark hair, dark eyes, and a darker skin than racial majority children. In contrast, the children displayed in the racial majority pictures had no such ‘‘foreign’’ attributes. The pictures had been previously shown to be perceived as ‘‘foreign’’ by German preservice teachers (Glock, Kneer, et al., 2013). The Chinese pictographs implemented in the AMP measure were downloaded from the Inquisit Millisecond Task Library (Millisecond Software). The primes in the AMP measure were the same pictures of male students used in the affective priming task and the IAT. We compiled a demographic questionnaire assessing participants’ age, gender, and teaching experience. Procedure Participants gave informed consent and were seated in front of the computer screen. The three different implicit attitudes measures appeared at random. In the AMP, each trial consisted of a picture followed by a Chinese pictograph. Participants’ task was to rate the pleasantness of the Chinese pictograph. Instructions appeared on the computer screen asking participants to evaluate the Chinese pictographs as pleasant by pressing the ‘‘E’’ key or as unpleasant by pressing the ‘‘I’’ key. Before the beginning
57
of the experimental procedure, participants were informed that the aim of this study was to assess their intuitive judgments of the pictographs and that these pictographs followed a picture. They were warned that the pictures might influence their evaluations but that they should avoid such an influence (Payne et al., 2007). Two different trial types were implemented in the task. One trial type consisted of a picture of a student from a racial minority, which was presented for 75 ms and followed by a Chinese pictograph, which was masked after 200 ms. A second trial type consisted of a picture of a student from a racial majority followed by a Chinese pictograph. The participants’ task was always to rate the pleasantness of the Chinese pictograph, independent of the trial type. Participants performed 10 practice trials before the 52 experimental trials were run in a random sequence. In the first phase of the IAT, participants were asked to categorize the pictures of the students into the categories indicating ‘‘from racial majority’’ by pressing the ‘‘I’’ key and ‘‘from racial minority’’ by pressing the ‘‘E’’ key of the keyboard. In the second phase, participants were presented with positive and negative adjectives and given the task of indicating the word valences by pressing the ‘‘I’’ key for positive and the ‘‘E’’ key for negative words. In the third phase, the two tasks were combined. Participants were asked to press the ‘‘I’’ key in order to indicate ‘‘from racial majority’’ and positive words and the ‘‘E’’ key in order to indicate ‘‘from racial minority’’ and negative words. Participants underwent 40 practice trials in order to become familiar with this combination before the 80 experimental trials were run. In the fourth phase, the combination of the keys was switched. Participants were now asked to press the ‘‘I’’ key for negative and the ‘‘E’’ key for positive words. In the fifth phase, this new assignment of the keys was combined with the pictures. Now, participants were asked to respond with the ‘‘I’’ to negative words and pictures from racial majorities and with the ‘‘E’’ key to positive words and pictures from racial minorities. After 40 practice trials, participants again underwent 80 experimental trials. Half of the participants were presented with the trials combining positive words with racial majority and negative words with racial minority pictures first. The other half of the participants received the trials combining positive words with racial minority and negative words with racial majority picture first. In the affective priming task, participants were informed that they would be presented with pictures followed by words and that their task was to indicate the valence of the words by pressing the ‘‘I’’ key to indicate positive and the ‘‘E’’ key to indicate negative word valences. Four different trials were implemented in this task. Pictures of racial minority students were followed either by positive or negative words, as were racial majority pictures. Participants underwent 10 practice trials, followed by 80 experimental trials. After completing the computerized tasks, participants filled in the demographic questionnaire and were thanked and debriefed. Results Affect misattribution procedure The sum of participants’ pleasant responses was calculated for racial majority and for racial minority pictures. We then divided the sum by the number of possible responses. The resulting value reflected the proportion of pleasant responses, with higher values indicating more positive implicit attitudes (Payne et al., 2007). The scores were submitted to a dependent t-test, t(64) = 4.40, d = 0.55, p < .001, which showed that participants more frequently evaluated the Chinese pictographs as pleasant after racial majority (M = 0.56, SD = 0.22) than after racial minority pictures (M = 0.41, SD = 0.20). Since we had no neutral primes included in the study,
58
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
we compared the frequency of pleasant responses to chance responding (50% of the trials). These single-sample t-tests revealed that participants’ frequency of pleasant responses after racial majority pictures was higher than chance, t(64) = 2.37, d = 0.27, p < .05, whilst the frequency of pleasant responses after minority pictures was lower than chance, t(64) = 3.57, d = 0.45, p < .001. To estimate the internal consistency of the AMP, we randomly selected 20 trials following racial majority pictures and 20 trials following racial minority pictures. We matched majority/minority trials using their order of selection and calculated difference scores for each trial pair. Cronbach’s a computed with the resulting 20 scores was .77. Implicit association test Following the suggestions of Greenwald et al. (1998), we replaced all latencies under 300 ms with 300 ms and above 3000 ms with 3000 ms and included all error trials. This procedure corrected the data for inattention and anticipation (Greenwald et al., 1998). Following the suggestions of Greenwald, Nosek, and Banaji (2003), the practice trials were included in calculating the D measure. First, we calculated the response latencies for trials combining racial majority pictures with positive words and racial minority pictures with negative words, and for trials combining racial majority pictures with negative words and racial minority pictures with positive words. Second, we calculated the D measure as the difference between the response latencies of the two different trial combinations divided by the standard deviation computed from the scores of the two combinations, D measure = 0.93, SD = 0.96. The latencies were submitted to a dependent t-test, t(64) = 7.79, d = 0.97, p < 001. Participants responded more slowly in trials combining racial minority pictures with positive words (M = 1039.99, SD = 221.35) than in trials combining racial minority pictures with negative words (M = 862.26, SD = 204.03). To estimate the internal consistency of the IAT, we randomly selected 20 trials combining racial majority pictures with positive words and 20 trials combining racial minority pictures with positive words. Using their selection order, we matched the two different trial types and calculated the IAT-scores for each of the 20 trial pairs. Cronbach’s a was .59. Affective priming task Response latencies were screened for erroneous classifications and latencies under 250 ms and above 1500 ms (Hermans, Baeyens, Lamote, Spruyt, & Eelen, 2005). These responses were excluded from further analysis (10% of all trials) in order to ensure that only correct and automatic responses were included in the data set. Response latencies were submitted to a 2 2 repeated measures ANOVA with picture (racial majority vs. racial minority) and word (positive vs. negative) as factors. ANOVA revealed a main effect of word, F(1, 64) = 16.03, p < .001, h2p ¼ 0:20, indicating that participants responded faster to positive (M = 716.63, SD = 118.61) than to negative words (M = 747.36, SD = 107.22). ANOVA also yielded a significant interaction, F(1, 64) = 13.08, p < .001, h2p ¼ 0:17. Participants responded more quickly to positive words (M = 703.39, SD = 121.39) than to negative words (M = 757.53, SD = 118.12) following racial majority pictures, t(64) = 5.57, d = 0.69, p < .001. Following racial minority pictures, there was no difference between positive (M = 729.86, SD = 125.40) and negative words (M = 737.19, SD = 106.90), t(64) = 0.71, d = 0.09, p = .48. Participants responded more quickly to positive words following racial majority than racial minority pictures, t(64) = 3.13, d = 0.39, p < .01, and more quickly to negative following racial minority than racial majority pictures, t(64) = 2.37, d = 0.29,
Table 1 Effect sizes d of each implicit measure. Implicit measure
IAT
AMP
APT (interaction)
Effect size d
0.93
0.55
0.91
p < .05. To determine the internal consistency of the affective priming task, we randomly ordered the four types of trials to create clusters of four response latencies (one for each trial type). Within each cluster, we first calculated two difference scores by subtracting the response latencies for positive words from the latencies for negative words following either racial majority or racial minority pictures. Then we subtracted the difference score for racial majority pictures from the difference score for racial minority pictures. The resulting 20 scores were used to compute Cronbach’s a = .42. Table 1 presents the effect size d of each measure. In a last step, we calculated the correlations between the three implicit measures. Thus, we transformed the values of each implicit measure into one single score. We calculated (1) the AMP score by subtracting the responses for racial majority from the responses for racial minority pictures. Positive values on this score reflected positive attitudes toward racial minorities (range: .88–.12). We calculated (2) the IAT score by subtracting the incompatible trials from the compatible trials. Positive values on this IAT score reflected positive attitudes toward racial minorities (range: 599.82– 313.80). We calculated the priming difference score using the same procedure as described for the calculation of the internal consistency of the affective priming task, except that we now relied on the means of each trial combination. Positive values on this priming difference score indicated positive attitudes toward racial minorities (range: 169.90–346.32). Intercorrelations are displayed in Table 2. The different implicit attitudes measures showed no significant correlations.
Discussion The AMP as well as the IAT revealed negative implicit attitudes toward racial minority. In the AMP, this was indicated by a higher proportion of pleasant responses as well as by the finding that the frequency of pleasant responses after racial minority pictures was lower than by chance. In the IAT, this can be concluded from the faster responses in compatible trials combining racial majority students with positive and racial minority students with negative words than in incompatible trials. However, a closer inspection of the results of the affective priming task revealed that participants showed positivity toward racial majority students rather than negativity toward racial minority students. This was indicated by faster responses to positive than to negative words following racial majority students, which is also in line with previous research (Glock, Kneer, et al., 2013). The affective priming task allows us to draw conclusions about the implicit attitudes toward racial minority as well as toward racial majority students (De Houwer, Teige-Mocigemba, Spruyt, & Moors, 2009), while the IAT only provides a single comparative measure indicating the difference in attitudes toward these two groups. Moreover, although we had no neutral primes included in the AMP measure, we might Table 2 Intercorrelations and p-values in parentheses between the three implicit attitudes measures.
AMP IAT Note: *p < .05.
IAT
Affective priming
.07 (.57)
.06 (.64) .09 (.47)
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
nevertheless be able to disentangle implicit attitudes toward racial minority students from those toward racial majority students (Dunham & Emory, 2014). Thus, from the AMP measure and the comparison of the frequency of pleasant responses to chance, we can cautiously conclude that the lower proportion of pleasant responses is due to negative implicit attitudes toward racial minority students as well as to positive attitudes toward racial majority students. These results partially support and partially stand in contrast to the findings of the affective priming measure. Hence, future research is needed to investigate why these two implicit measures revealed different implicit attitudes toward racial minority students. Including neutral primes in both measures in future research might help us gain a deeper understanding of the discrepancies in our results. One difference between the AMP and the affective priming task is that the former relies on ratings while the latter relies on response latencies (Payne et al., 2007). Compared to the affective priming task, the AMP is superior in reliability and in predictive validity (Payne, 2009). Nevertheless, on order to disentangle attitudes toward racial minority students from those toward majority students, future research should employ additional implicit reliable measures such as the Single-Category IAT (Karpinski & Steinman, 2006) or the Extrinsic Affective Simon Task EAST (De Houwer, 2003), which also allows to disentangle attitudes toward racial minority from those toward majority. This could shed light on the question of whether implicit positivity toward racial majority, implicit negativity toward racial minority students, or both caused the effects. The three implicit attitudes measures showed no substantial correlations, which is not entirely surprising, given that correlations between the IAT and the affective priming task are generally low (Olson & Fazio, 2003; Payne, Govorun, et al., 2008). The IAT measures attitudes toward categories while the affective priming task measures attitudes toward specific exemplars which are then averaged into the categories (Olson & Fazio, 2003). Thus, the correlation between the two implicit measures increases when the affective priming task also requires a categorization (Olson & Fazio, 2003). Although the AMP as well as the affective priming task are priming measures, the AMP relies on pleasant and unpleasant responses, while the affective priming tasks relies on response latencies, which might partially explain the low correlations. A study employing the same three implicit measures of attitudes also reported no significant correlations between the AMP, the IAT, and the affective priming task (Payne, Govorun, et al., 2008). Interestingly, this study also implemented consistent stimulus materials into all three measures, thus, the low correlations did not stem from variations in the material used. The low correlations we found between different implicit measures are in line with previous research (e.g., Olson & Fazio, 2003; Sherman et al., 2003) and might – to some extent – reflect measurement error (Cunningham, Preacher, & Banaji, 2001; De Houwer et al., 2009) as well as the different structure underlying each implicit measure (De Houwer, 2009). The reliability values of the three measures support this assumption. Although the AMP showed rather good reliability, the IAT showed moderate and the affective priming task low reliability values. Thus, the low correlations might indeed partially stem from measurement error, however, it also might be that – besides the different structure of the implicit measures – they tap different facets of implicit attitudes reflecting the complexity of implicit attitudes (Bosson, Swann, & Pennebaker, 2000). Nonetheless, our results have implications for teachers’ and preservice teachers’ judgments. Although preservice teachers implicit attitudes did not differ from the general population (Nosek, Banaji, & Greenwald, 2002), their attitudes are of particular interest. These educators will assign grades and decide about the school career of students. Thus, preservice teachers’ attitudes could play a pivotal role for the educational career of the students.
59
Given the negative racial bias found in the IAT and the AMP, all three implicit measures imply more favorable preservice teachers’ attitudes, judgments, and behaviors toward racial majority students. This implicit racial bias might contribute to disadvantages experienced by racial minority students. Since teachers often compare the academic achievement of a student with the achievement of his or her peers in the class (Su¨dkamp et al., 2012), the implicit racial bias may play a pivotal role when judging a racial minority student in comparison to a racial majority student. This assumption is in line with previous research showing that, compared to a racial majority student displaying the same academic profile, racial minority students are judged less favorably (Glock, Krolak-Schwerdt, et al., 2013; Glock & Krolak-Schwerdt, 2013). Notwithstanding the implicit racial bias we found in this study, implicit attitudes have been shown to be malleable (Blair, 2002; Dasgupta & Greenwald, 2001) with the attitude object stored in a multifaceted manner (Gawronski & Bodenhausen, 2006). The context in which implicit attitudes are activated may play a large role in determining which automatic evaluation is activated (Fazio, 2007). In our study, we employed no specific cue to activate a particular evaluation, meaning that the resulting measures can be seen as ‘‘baseline’’ attitudes. Previous research has shown that confronting participants with positive Black exemplars resulted in more positive implicit attitudes toward Black individuals than confronting participants with a negative Black exemplar (Dasgupta & Greenwald, 2001). In this vein, implicit racial attitudes can be changed through diversity education (Rudman, Ashmore, & Gary, 2001). Therefore, it might be valuable for future research to investigate whether confronting preservice teachers with an expectation-confirming racial minority student can turn those ambivalent implicit attitudes into more negative ones and whether more positive implicit attitudes can be induced by presenting participants with an expectation-disconfirming racial minority student. Previous research has shown that explicitly assessed stereotypical beliefs are more positive after preservice teachers have been confronted with an expectation-disconfirming racial minority student (Glock & Krolak-Schwerdt, 2013). Moreover, it has also been shown that racial minority students are only judged less favorably when they are prototypical members of this social group (Glock & Krolak-Schwerdt, 2013; Parks & Kennedy, 2007). Therefore, when preservice teachers enter the classroom and are confronted with an expectation-confirming racial minority student, activated negative implicit attitudes might affect their judgments, possibly leading to less favorable judgments even though this students’ academic performance is equal to that of racial majority students. Our study has some limiting aspects that should be kept in mind. As the focus of our study was to employ different implicit measures of attitudes in order to investigate the nature of implicit attitudes toward racial minority students, we did not assess explicit attitudes. Since explicit attitudes measures rely on selfreports, they may – to a substantial extent – reflect social norms and social desirability rather than explicit attitudes. However, future research should also employ an explicit attitudes measure, such as the Teacher Cultural Beliefs Scale (Hachfeld et al., 2011). This would shed further light on possible differences between implicit and explicit attitudes, thus providing a deeper understanding of teachers’ attitudes. We did not assess actual behavior, judgments, or even behavioral intentions. Thus, future research should relate implicit attitudes to actual behavior, judgments, or behavioral intentions in order to investigate whether implicit attitudes contribute to judgment biases. Since implicit attitudes are automatically activated by the mere presence of the attitude object (Fazio, 2001), it seems plausible that preservice teachers’ behavior might be affected by implicit attitudes. Particularly, given that people are
60
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61
often not aware of this influence (Asendorpf et al., 2002) and that implicit attitudes exhibit a greater influence when cognitive resources are limited (Fazio, 1990; Olson & Fazio, 2009), classroom behavior is likely to be guided by implicit attitudes. No experienced teachers were included in our sample. Thus, our results are limited to preservice teachers and we cannot draw conclusions about the nature of experienced teachers’ implicit attitudes toward racial minority students. One study exploring implicit attitudes toward racial minority students found slightly negative implicit attitudes, which were related to the academic achievement of the students assessed via standardized tests (Van den Bergh et al., 2010). Racial minority students instructed by teachers with more negative implicit attitudes achieved lower test scores than racial minority students instructed by teachers with more positive implicit attitudes (Van den Bergh et al., 2010). This research emphasizes the pivotal role of implicit attitudes in the classroom routine – and consequently the vital role of research on experienced teachers’ implicit attitudes toward racial minority students. Notwithstanding the limitations of our study, the employment of three different implicit attitudes measures provided a consistent pattern of implicit attitudes that revealed more positive attitudes toward racial majority than toward racial minority students. Keeping in mind that preservice teachers are future teachers, our study provides valuable insights into the nature of their implicit attitudes, which could be used to implement specific training courses early during their academic studies in order to overcome implicit positivity toward racial majority students and help ensure the fair treatment of all students.
Author note We would like to thank Carrie Kovacs for proof reading the manuscript.
References Ansalone, G. (2001). Schooling, tracking, and inequality. Journal of Children and Poverty, 7, 33–47. http://dx.doi.org/10.1080/10796120120038028 Ansalone, G., & Biafora, F. (2004). Elementary school teachers’ perceptions and attitudes to the educational structure of tracking. Education, 125, 249–260. Asendorpf, J. B., Banse, R., & Mu¨cke, D. (2002). Double dissociation between implicit and explicit personality self-concept: The case of shy behavior. Journal of Personality and Social Psychology, 83, 380–393. http://dx.doi.org/10.1037//0022-3514.83.2.380 Blair, I. (2002). The malleability of automatic stereotypes and prejudice. Personality and Social Psychology Review, 6, 242–261. http://dx.doi.org/10.1207/S15327957 PSPR0603_8 Bosson, J. K., Swann, W. B., & Pennebaker, J. W. (2000). Stalking the perfect measure of implicit self-esteem: The blind men and the elephant revisited? Journal of Personality and Social Psychology, 79, 631–643. Caro, D. H., Lenkeit, J., Lehmann, R., & Schwippert, K. (2009). The role of academic achievement growth in school track recommendations. Studies in Educational Evaluation, 35, 183–192. http://dx.doi.org/10.1016/j.stueduc.2009.12.002 Coneus, K., Gernandt, J., & Saam, M. (2009). Noncognitive skills, school achievements and educational dropout. ZEW Discussion Papers, No. 09-019. Cunningham, W. A., Preacher, K. J., & Banaji, M. R. (2001). Implicit attitude measures: Consistency, stability, and convergent validity. Psychological Science, 12, 163–170. Dasgupta, N., & Greenwald, A. G. (2001). On the malleability of automatic attitudes: Combating automatic prejudice with images of admired and disliked individuals. Journal of Personality and Social Psychology, 81, 800–814. http://dx.doi.org/10.1037/ 0022-3514.81.5.800 De Houwer, J. (2003). The extrinsic affective Simon task. Experimental Psychology, 50, 77–85. http://dx.doi.org/10.1027//1618-3169.50.2.77 De Houwer, J. (2006). What are implicit measures and why are we using them? In R. W. Wiers & A. W. Stacy (Eds.), Handbook of implicit cognition and addiction (pp. 11–28). Thousand Oaks, CA: Sage Publisher. De Houwer, J. (2009). Comparing measures of attitudes at the functional and procedural level: Analysis and implications. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 361–390). New York, NY: Psychology Press. De Houwer, J., Geldof, T., De Bruycker, E., & De Bruycher, E. (2005). The implicit association test as a general measure of similarity. Canadian Journal of Experimental Psychology, 59, 228–239. http://dx.doi.org/10.1037/h0087478
De Houwer, J., Teige-Mocigemba, S., Spruyt, A., & Moors, A. (2009). Implicit measures: A normative analysis and review. Psychological Bulletin, 135, 347–368. http:// dx.doi.org/10.1037/a0014211 Dee, T. S. (2005). A teacher like me: Does race, ethnicity, or gender matter? The American Economic Review, 95, 158–165. http://dx.doi.org/10.1257/00028280 5774670446 Destatis (2012). Ausla¨ndische Bevo¨lkerung [foreign population]. Available from: https://www.destatis.de/DE/ZahlenFakten/GesellschaftStaat/Bevoelkerung/ MigrationIntegration/AuslaendischeBevolkerung/Tabellen/Staatsange hoerigkeitJahre.html Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18. http:// dx.doi.org/10.1037//0022-3514.56.1.5 Dovidio, J. F., Brigham, J. C., Johnson, B. S., & Gaertner, S. L. (1996). Stereotyping, prejudice, and discrimination: Another look. In C. N. Macrae, M. Hewstone, & C. Stangor (Eds.), Stereotypes and stereotyping (pp. 276–319). New York: Guilford Press. Dovidio, J. F., & Fazio, R. H. (1992). New technologies for the direct and indirect assessment of attitudes. In J. M. Tanur (Ed.), Questions about questions: Inquiries into the cognitive bases of surveys (pp. 204–237). New York, NY: Russel Sage Foundation. Dunham, Y., & Emory, J. (2014). Of affect and ambiguity: The emergence of preference for arbitrary ingroups. Journal of Social Issues, 70, 81–98. http://dx.doi.org/10.1111/ josi.12048 Euwals, R., Dagevos, J., Gijsberts, M., & Roodenburg, H. (2007). Immigration, integration, and the labour market: Turkish immigrants in Germany and the Netherlands. Discussion Paper no. 2677. Bonn: IZA. Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (23, pp. 75–109). New York, NY: Academic Press. Fazio, R. H. (2001). On the automatic activation of associated evaluations: An overview. Cognition and Emotion, 15, 115–141. http://dx.doi.org/10.1080/026999300 4200024 Fazio, R. H. (2007). Attitudes as object-evaluation associations of varying strength. Social Cognition, 25, 603–637. http://dx.doi.org/10.1521/soco.2007.25.5.603 Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology, 69, 1013–1027. http://dx.doi.org/ 10.1037/0022-3514.69.6.1013 Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229–238. http://dx.doi.org/10.1037//0022-3514.50.2.229 Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change. Psychological Bulletin, 132, 692–731. http://dx.doi.org/10.1037/0033-2909.132.5.692 Gawronski, B., Strack, F., & Bodenhausen, G. V. (2009). Attitudes and cognitive consistency. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 85–117). New York, NY: Psychology Press. Glock, S., Kneer, J., & Kovacs, C. (2013). Preservice teachers’ implicit attitudes toward students with and without immigration background: A pilot study. Studies in Educational Evaluation, 39, 204–210. http://dx.doi.org/10.1016/j.stueduc.2013. 09.003 Glock, S., & Kovacs, C. (2013). Educational psychology: Using insights from implicit attitude measures. Educational Psychology Review, 25, 503–522. http://dx.doi.org/ 10.1007/s10648-013-9241-3 Glock, S., & Krolak-Schwerdt, S. (2013). Does nationality matter? The impact of stereotypical expectations on student teachers’ judgments. Social Psychology of Education, 16, 111–127. http://dx.doi.org/10.1007/s11218-012-9197-z Glock, S., Krolak-Schwerdt, S., Klapproth, F., & Bo¨hmer, M. (2013). Beyond judgment bias: How students’ ethnicity and academic profile consistency influence teachers’ tracking judgments. Social Psychology of Education, 16, 555–573. http://dx.doi.org/ 10.1007/s11218-013-9227-5 Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74, 1464–1480. http://dx.doi.org/10.1037/0022-3514.74.6. 1464 Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the implicit association test I: An improved scoring algorithm. Journal of Personality and Social Psychology, 85, 197–216. http://dx.doi.org/10.1037/0022-3514.85.2. 197 Hachfeld, A., Hahn, A., Schroeder, S., Anders, Y., Stanat, P., & Kunter, M. (2011). Assessing teachers’ multicultural and egalitarian beliefs: The teacher cultural beliefs scale. Teaching and Teacher Education, 27, 986–996. http://dx.doi.org/10. 1016/j.tate.2011.04.006 Haycock, K. (2001). Closing the achievement gap. Educational Leadership, 58, 6–11. Hermans, D., Baeyens, F., Lamote, S., Spruyt, A., & Eelen, P. (2005). Affective priming as an indirect measure of food preferences acquired through odor conditioning. Experimental Psychology, 52, 180–186. http://dx.doi.org/10.1027/1618-3169.52.3. 180 Hofmann, W., Gschwendner, T., Nosek, B. A., & Schmitt, M. (2005). What moderates implicit—Explicit consistency? European Review of Social Psychology, 16, 335–390. http://dx.doi.org/10.1080/10463280500443228 Jussim, L., & Harber, K. D. (2005). Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9, 131–155. http://dx.doi.org/10.1207/s15327957 pspr0902_3
S. Glock, J. Karbach / Studies in Educational Evaluation 45 (2015) 55–61 Karpinski, A., & Hilton, J. L. (2001). Attitudes and the implicit association test. Journal of Personality and Social Psychology, 81, 774–788. http://dx.doi.org/10.1037//00223514.81.5.774 Karpinski, A., & Steinman, R. B. (2006). The single category implicit association test as a measure of implicit social cognition. Journal of Personality and Social Psychology, 91, 16–32. http://dx.doi.org/10.1037/0022-3514.91.1.16 Kristen, C., & Granato, N. (2007). The educational attainment of the second generation in Germany: Social origins and ethnic inequality. Ethnicities, 7, 343–366. http:// dx.doi.org/10.1177/1468796807080233 Lee, J. (2002). Racial and ethnic achievement gap trends: Reversing the progress toward equity. Educational Researcher, 31, 3–12. http://dx.doi.org/10.3102/ 0013189X031001003 Lucas, S. R. (2001). Effectively maintained inequality: Education transitions, track mobility, and social background effects. American Journal of Sociology, 106, 1642–1690. http://dx.doi.org/10.1086/321300 McCombs, R. C., & Gay, J. (1988). Effects of race, class, and IQ information on judgments of parochial grade school teachers. The Journal of Social Psychology, 128, 647–652. http://dx.doi.org/10.1080/00224545.1988.9922918 Millisecond Software. (n.d.). Inquisit task library. Available from http://www. millisecond.com/download/library/AMP/. Moors, A., & De Houwer, J. (2006). Automaticity: A theoretical and conceptual analysis. Psychological Bulletin, 132, 297–326. http://dx.doi.org/10.1037/0033-2909.132.2. 297 Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). Harvesting implicit group attitudes and beliefs from a demonstration web site. Group Dynamics: Theory, Research, and Practice, 6, 101–115. http://dx.doi.org/10.1037//1089-2699.6.1.101 Oakes, J. (2005). Keeping track: How schools structure inequality (2nd ed.). New Haven, CT: Yale University Press. Olson, M. A., & Fazio, R. H. (2003). Relations between implicit measures of prejudice: What are we measuring? Psychological Science, 14, 636–639. http://dx.doi.org/ 10.1046/j.0956-7976.2003.psci_1477.x Olson, M. A., & Fazio, R. H. (2009). Implicit and explicit measures of attitudes: The perspective of the MODE model. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures (pp. 19–63). New York, NY: Psychology Press. Parks, F. R., & Kennedy, J. H. (2007). The impact of race, physical attractiveness, and gender on education majors’ and teachers’ perceptions of student competence. Journal of Black Studies, 37, 936–943. http://dx.doi.org/10.1177/0021934705285955 Payne, B. K. (2009). Attitude misattribution: Implications for attitude measurement and the implicit-explicit relationship. In R. E. Petty, R. H. Fazio, & P. Brin˜ol (Eds.), Attitudes: Insights from the new implicit measures. Psychology Press: New York, NY. Payne, B. K., Burkley, M., & Stokes, M. B. (2008). Why do implicit and explicit attitude tests diverge? The role of structural fit. Journal of Personality and Social Psychology, 94, 16–31. http://dx.doi.org/10.1037/0022-3514.94.1.16 Payne, B. K., Cheng, S. M., Govorun, O., & Stewart, B. D. (2005). An inkblot for attitudes: Affect misattribution as implicit measurement. Journal of Personality and Social Psychology, 89, 277–293. http://dx.doi.org/10.1037/0022-3514.89.3.277 Payne, B. K., Govorun, O., & Arbuckle, N. L. (2008). Automatic attitudes and alcohol: Does implicit liking predict drinking? Cognition & Emotion, 22, 238–271. http:// dx.doi.org/10.1080/02699930701357394 Payne, B. K., McClernon, F. J., & Dobbins, I. G. (2007). Automatic affective responses to smoking cues. Experimental and Clinical Psychopharmacology, 15, 400–409. http:// dx.doi.org/10.1037/1064-1297.15.4.400 Rakic´, T., Steffens, M. C., & Mummendey, A. (2011). Blinded by the accent! The minor role of looks in ethnic categorization Journal of Personality and Social Psychology, 100, 16–29. http://dx.doi.org/10.1037/a0021522
61
Robinson, M. D., Meier, B. P., Zetocha, K. J., & McCaul, K. D. (2005). Smoking and the implicit association test: When the contrast category determines the theoretical conclusions. Basic and Applied Social Psychology, 27, 201–212. http://dx.doi.org/ 10.1207/s15324834basp2703_2 Rudman, L. A., Ashmore, R. D., & Gary, M. L. (2001). Unlearning, automatic biases: The malleability of implicit prejudice and stereotypes. Journal of Personality and Social Psychology, 81, 856–868. http://dx.doi.org/10.1037//0022-3514.81.5.856 Sanbonmatsu, D. M., & Fazio, R. H. (1990). The role of attitudes in memory-based decision making. Journal of Personality and Social Psychology, 59, 614–622. http:// dx.doi.org/10.1037//0022-3514.59.4.614 Santavirta, N., Solovieva, S., & Theorell, T. (2007). The association between job strain and emotional exhaustion in a cohort of 1.028 Finnish teachers. British Journal of Educational Psychology, 77, 213–228. http://dx.doi.org/10.1348/000709905X92045 Schnabel, K., Asendorpf, J. B., & Greenwald, A. G. (2008). Assessment of individual differences in implicit cognition. European Journal of Psychological Assessment, 24, 210–217. http://dx.doi.org/10.1027/1015-5759.24.4.210 Sherman, S. J., Rose, J. S., Koch, K., Presson, C. C., & Chassin, L. (2003). Implicit and explicit attitudes toward cigarette smoking: The effects of context and motivation. Journal of Social and Clinical Psychology, 22, 13–39. http://dx.doi.org/10.1521/ jscp.22.1.13.22766 Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220–247. http://dx.doi.org/ 10.1207/s15327957pspr0803_1 Su¨dkamp, A., Kaiser, J., & Mo¨ller, J. (2012). Accuracy of teachers’ judgments of students’ academic achievement: A meta-analysis. Journal of Educational Psychology, 104, 743–762. http://dx.doi.org/10.1037/a0027627 Tenenbaum, H. R., & Ruck, M. D. (2007). Are teachers’ expectations different for racial minority than for European American students? A meta-analysis. Journal of Educational Psychology, 99, 253–273. http://dx.doi.org/10.1037/0022-0663.99.2.253 Van den Bergh, L., Denessen, E., Hornstra, L., Voeten, M., & Holland, R. W. (2010). The implicit prejudiced attitudes of teachers: Relations to teacher expectations and the ethnic achievement gap. American Educational Research Journal, 47, 497–527. http://dx.doi.org/10.3102/0002831209353594 Van Dick, R., & Wagner, U. (2001). Stress and strain in teaching: A structural equation approach. British Journal of Educational Psychology, 71, 243–259. http://dx.doi.org/ 10.1348/000709901158505 Wilson, T. D., Lindsey, S., & Schooler, T. Y. (2000). A model of dual attitudes. Psychological Review, 107, 101–126. http://dx.doi.org/10.1037//0033-295X.107.1.101
Sabine Glock is an educational psychologist holding a PhD in psychology from the University of Saarland. Since 2014, she is a member of the Institute for Educational Research in the School of Education at the Bergische Universita¨t Wuppertal, Germany. She is concerned with educational decision making and factors influencing educational decisions. Her main research interest is implicit cognition, in particular implicit attitudes toward students with ethnic/racial minority background and how they contribute to teacher judgment
Julia Karbach is a cognitive developmental psychologist holding a PhD in psychology from the University of Saarland. She currently serves as professor at Goethe University Frankfurt, Germany. Her main research interests include cognitive development and plasticity across the lifespan as well as teacher education. Her work has been published in leading peer-reviewed international journals.