An examination of the relationship between social self-efficacy and personal growth initiative in international context

International Journal of Intercultural Relations 61 (2017) 88–96

Contents lists available at ScienceDirect

International Journal of Intercultural Relations journal homepage: www.elsevier.com/locate/ijintrel

An examination of the relationship between social self-efficacy and personal growth initiative in international context☆

MARK

⁎

Elif Merve Çankaya , Xin Dong, Jeffrey Liew Texas A & M University, Department of Educational Psychology, College Station, TX 77843-4222, United States

AR TI CLE I NF O

AB S T R A CT

Keywords: Personal growth initiative Social self-efficacy International students PGIS-II

The cross-cultural transitioning experience of international students has been viewed as a highimpact practice for student learning. International students experience both challenges and growth opportunities when they encounter new peoples and environments. However, limited research exists on the student factors that predict benefits, growth or resilience for international students through their cross-cultural transitioning experience. This study investigated social selfefficacy as a predictor of personal growth initiative amongst international students, above and beyond demographic and linguistic factors. In addition, the factorial structure of Personal Growth Initiative Scale-II was examined. Our results indicated that the perceived ability to successfully negotiate social situations and produce positive social interactions predict greater personal growth initiative and its dimensions regardless of students’ language proficiency. Moreover, results from the confirmatory factor analysis (CFA) provided further support for the first-order four-factor structure of the PGIS-II. Results have implications for improving international student integration, because efforts to support students’ social self-efficacy are linked to their personal growth initiative.

Introduction International education has the potential to present students with high-impact learning and growth opportunities. However, international students who study beyond the borders of their home country encounter growth opportunities as well as challenges. Numerous studies have revealed the potential benefits of dealing with demanding life challenges (e.g., Tedeschi & Calhoun, 2004). Those who move to and live in places very different than from their native environments inevitably encounter challenging experiences that require major life adjustments. One example is international students, who seek a cross-cultural transitioning experience and undergo major life adjustments, which can be challenging as well as enriching. The vast majority of studies on the impacts of the international education and the cross-cultural transitioning experience of international students tend to focus on stress and coping or cross-cultural adaptability (e.g., Williams, 2005), and adopts a pathological view of international students and their experiences in comparison to the experiences of their domestic peers (Lillyman & Bennett, 2014; Moores & Popadiuk,2011). Such a skewed view has resulted in a negative depiction of international students and their experiences and ignores their growth initiative and resilience. By contrast, far fewer studies have explored international students’ experiences beyond the challenges of cross-cultural adaptability and integration, and those studies have revealed that international students successfully navigate difficulties in their new environment and utilize those difficult experiences as

☆ ⁎

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Corresponding author. E-mail address: [email protected] (E.M. Çankaya).

http://dx.doi.org/10.1016/j.ijintrel.2017.10.001 Received 29 November 2016; Received in revised form 28 September 2017; Accepted 9 October 2017 0147-1767/ © 2017 Elsevier Ltd. All rights reserved.

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

sources of personal growth (e.g., Moores & Popadiuk, 2011). However, the impact of cross-cultural transitioning experiences might differ for students, depending on their social self-efficacy and their motivation to pursue personal growth. Thus, the present study aims to balance the negatively skewed perspective of international students by exploring the neglected positive aspects of their experiences with a focus on social self-efficacy and personal growth initiative (PGI). Many theories recognize that continued personal growth throughout the life span is an indicator of healthy functioning (e.g., Robitschek, 1998). Individuals with high PGI are found to successfully manage stressors and challenges throughout their lives, report better adjustment to the environment (Yakunina, Weigold, & Weigold, 2013), have high levels of well-being (Robitschek & Keyes, 2009), and experience fewer psychological problems such as anxiety and depression (Robitschek & Kashubeck, 1999). PGI also plays an important role in international students’ cross-cultural transition process. For example, high levels of PGI and multicultural strength are predictors of fewer acculturation problems, and they lead to better adjustment to a new country (Yakunina, Weigold, Weigold, Hercegovac, & Elsayed, 2013). Given the fact that PGI is viewed as an indicator of positive functioning, healthy personality, and greater adjustment, investigating predictors of PGI could provide promising benefits by targeting ways to boost PGI. One construct proposed as relevant to PGI is selfefficacy, because PGI has been posited to reflect the cognitive components of self-efficacy “including beliefs, attitudes and values that support personal growth” (Robitschek, 1998, p. 184). Like PGI, self-efficacy has also been related to cross-cultural adjustment (Harrison, Chadwick, & Scales, 1996; Zhang & Goodson, 2011). Existing research studies have shown a link between generalized self-efficacy and PGI (e.g., Ogunyemi & Mabekoje, 2007; Sharma & Rani, 2013; Weigold et al., 2014). More specifically, self-efficacy was a predictor of PGI in a sample of graduate students in India (Sharma & Rani, 2013) and in a sample of university students in Nigeria (Ogunyemi & Mabekoje, 2007). A notable gap in the these studies, however, was the inconsistent strength of the relations found between self-efficacy and PGI, which ranged from small to large. In addition, these studies reporting self-efficacy as a predictor of PGI either utilized the older version of the scale used to assess PGI or lack evidence of the validity of the factorial structure of the recently redesigned Personal Growth Initiative Scale-II (PGIS-II). PGI is conceived of as a multidimensional construct with distinct cognitive and behavioral dimensions. The original instrument developed to assess PGI was a one-dimensional nine-item scale, and the use of the original measure was limited in assessing the multiple components (cognitive and behavioral) of the construct and fully capturing the complex growth process. Thus, a multidimensional instrument (PGIS-II) with cognitive and behavioral components was reconstructed to overcome the limitations of the initial measure (PGIS-II, Robitschek et al., 2012). In the development study, three factorial structures for the new instrument (single factor, first-order four-factor, and second-order four-factor) were examined. The results revealed the first-order four-factorial structure as the best fitting model, supporting the existence of two cognitive (readiness for change and planfulness) and two behavioral (using resources and intentional behavior) dimensions that compose PGI (Robitschek et al., 2012). Prior empirical research utilized the redesigned instrument evaluated PGI as a total score obtained through averaging the scores of each factor (e.g., Sharma & Rani, 2013; Thoen & Robitschek, 2013; Yakunina, Weigold, Weigold, Hercegovac et al., 2013), as recommended by Robitschek et al., 2012. However, considering the established first-order four-factor structure of the instrument, evidence lacking to support the use of a general PGI score being greater than its four dimensions (Freitas, Damásio, Tobo, Kamei, & Koller, 2016). A study conducted in a sample of international students also suggested the utilization of the first-order solution when evaluating PGI (Yakunina, Weigold, Weigold et al., 2013), prompting the researchers to call for further research to validate the factorial structure of the instrument, specifically among samples of international group of peoples. Moreover, the unstable factorial structure revealed with the cross-cultural use of the scale (e.g., Bhattacharya & Mehrotra, 2014 in India), and the addition of error covariances among the scale items to obtain an adequate fit (e.g., Freitas, Damásio, Kamei, Tobo, Koller, Robistcheck, 2017; Weigold et al., 2014; Yakunina, Weigold, Weigold et al., 2013) necessitated further evaluation of the instrument. In light of calls by researchers in the field to further examine the structure of the PGIS-II (e.g., Freitas et al., 2016; Yakunina, Weigold, Weigold et al., 2013), we examined the proposed multidimensional factorial structure of the instrument, and explored the plausibility of the second-order factorial structure of PGIS-II to use an overall PGI in evaluating the construct. Moreover, given that many of the difficulties associated with international student’s cross-cultural adjustment are related to changes in social norms and cultural background, we argue that social self-efficacy (more so than generalized self-efficacy) would play an important role in the adjustment process. Thus, this study explored whether social self-efficacy predicts PGI, above and beyond demographic and linguistic factors, in a sample of international students. Literature review Personal growth initiative The urge for personal growth and enhancement throughout life can be considered one of the universal human motivation or needs. Personal growth initiative (PGI) represents an active and intentional engagement in the process of enhancing and developing one’s self across multiple life domains (Robitschek et al., 2012). Personal growth can occur in one of three different ways: developmental, environmental, and intentional. While growth resulting from developmental and environmental factors occurs without the intention or awareness of the individual, growth resulting from intentional processes requires one to be fully aware of, and committed to, the cognitive and behavioral aspects of changing one’s self (Robitschek, 1998). The salient aspect of the construct of PGI, which differs from other conceptualizations of the term “personal growth” in the literature, is the intentionality with which on engages the self-change process. In support of this difference, intentional engagement in the growth process has been found to be positively 89

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

correlated with, whereas a lack of intentionality is negatively correlated with, an individual’s psychological well-being (Robitschek, 1999). PGI is conceptualized of as a multidimensional construct that consists of distinct cognitive and behavioral dimensions (Robitschek et al., 2012). Specifically, PGI comprises the cognitive aspects of Bandura’s self-efficacy theory: a person’s knowledge of the cognitive processes of the ongoing change, general knowledge of the whole change process, and an appreciation of self-development. The behavioral dimension of PGI encompasses an individual’s making efforts to create opportunities for growth, benefitting from those opportunities in order to further self-development and growth, and adopting a course of action to initiate the growth process (Robitschek et al., 2012). PGIS-II was developed and validated in a sample of European Americans to assess PGI. The instrument consists of four subscales, two of which (planfulness and readiness for change) represent the cognitive aspect of PGI theory, while the remaining two subscales (using resources and intentional behavior) refer to the behavioral aspect. To provide further evidence of the multidimensional structure of PGIS-II, researchers examined the three proposed factorial structures (single factor, first-order four-factor, and second-order four-factor) for the PGIS-II tested in the initial study (see Robitschek et al., 2012). The result of the confirmatory factor analyses (CFA) revealed the first-order four-factor with correlated error terms as the best-fitting model in samples of non-native higher education students (Yakunina, Weigold, Weigold et al., 2013), and African American college students (Weigold et al., 2014). The multifaceted structure of the instrument was also examined in the adaptation studies into Turkish, Chinese and Portuguese languages (e.g., Freitas et al., 2017; Yalçın & Malkoç, 2013; Yang & Chang, 2014). The findings of these adaptation studies also lend support for the first-order four factorial structure of the instrument. Nonetheless, though the new scale was found to reflect the cognitive and behavioral aspect of the PGI construct, a different factorial model was observed with the cross-cultural use of the scale (i.e., Bhattacharya & Mehrotra, 2014 in India). The scientific effort given to examine the cross cultural relevance of PGI and the factorial structure of PGIS-II across various cultural groups is important, since PGI is proposed as a universal construct linked to several favorable outcomes. Based on PGI theory, high levels of PGI are related to psychologically healthy and well-adjusted human characteristics, regardless of the perceived distress in life. Individuals with high PGI view the challenging aspects of their experiences as an opportunity for personal growth, and they maintain a belief in their ability to cope successfully with challenges (Robitschek et. al., 2012). In support of this theoretical assumption, international students with high levels of PGI were found to adjust better to their new environment and to cope better with acculturative stress (Yakunina, Weigold, Weigold et al., 2013; Yakunina, Weigold, Weigold, Hercegovac et al., 2013). Social self-efficacy Self-efficacy is defined as the beliefs that one holds about one’s capacity to perform a specific task (Bandura, 1997). According to Bandura’s social cognitive theory, self-efficacy expectations and our beliefs concerning our competence in specific behavioral domains impact our choices, performance, and persistence in domain-related tasks. People with high self-efficacy put forth more effort when they face challenges, so higher efficacy beliefs are related to higher levels of general adjustment (Bandura, 1997). Bandura (1997) proposed that self-efficacy is domain-specific, which means that one’s beliefs concerning one’s competence in a certain behavioral domain have a great impact on one’s choices, performance, and persistence in this domain. Since the present study focused on international students’ social cultural adjustment, we decided to examine international students’ social self-efficacy, which refers to a person’s confidence in his or her ability to initiate and maintain social behaviors (Smith & Betz, 2000). Lin and Betz (2009) provided support for the role of social self-efficacy in adjustment in their study of Chinese international students. They found that students’ social self-efficacy was positively related to their English-language proficiency and the length of their residence in the U.S., and negatively related to acculturation stress (an important indicator of cross-cultural adjustment). The main goal of the present study was to examine the predictive relation between two constructs that are relevant to international students’ daily functioning: PGI and social self-efficacy. We hypothesized that PGI and social self-efficacy would be positively correlated, and that social self-efficacy would predict PGI, even after controlling for demographic and linguistic factors. Moreover, the multifaceted structure of the PGIS-II was examined. Consistent with prior research and theory, the first-order, four-factor structural model was hypothesized to be the best fit for our sample (Robitschek et al., 2012; Yakunina, Weigold, Weigold et al., 2013; Weigold et al., 2014). Methodology Participants The participants of this study consisted of 188 international graduate students studying at a large Midwestern university in the U.S. Fifty-three percent (n = 100) of the participants were male graduate students, and the remaining 47% (n = 88) were female graduate students. Forty-three percent (n = 80) of the participants were studying at the master’s level, and 57% (n = 108) at the doctoral level. Study participants’ ages ranged from 21 to 50 (M = 27.74 SD = 5.40). Instruments Demographic questionnaire A demographic questionnaire was developed to gather information about participants’ background, including the length of their residence in the U.S. and the participants’ self-rated English language proficiency, which was scored on a five-point Likert-scale from 90

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

1 (poor) to 5 (excellent). Personal growth initiative scale-II The Personal Growth Initiative Scale (PGIS-II, Robitschek et al., 2012) has 16 items grouped into four subscales: Planfulness, readiness for change, using resources and intentional behavior. The Planfulness subscale includes five items assessing the respondent’s ability to plan the process of personal growth. The Readiness for Change subscale is composed of four items measuring aspects of one’s preparedness to make specific changes in one’s self. The Using Resources subscale includes three items assessing the respondent’s ability to obtain assistance with the personal growth process from resources outside one’s self, and the Intentional Behavior subscale is composed of four items measuring participant’s conscious engagement in behaviors when growing. Items on the PGIS-II scale are rated on a six-point Likert-type scale, the values of the items on the scale ranging from 0 to 5. Similar to the results in the initial study conducted with college students and community samples by Robitschek et al. (2012), the alpha coefficient of the current study was 0.93 for the general score of PGI, and the alpha coefficients of the subscales ranged from 0.78 to 0.88. The scale of perceived social self-efficacy The Scale of Perceived Social Self-Efficacy (PSSE) (Smith & Betz, 2000) was used to measure social self-efficacy. This scale consists of 25 items assessing the respondents’ self-confidence in their ability to function in six areas of social tasks: making friends, pursuing romantic relationships, asserting themselves socially, performing in public situations, performing in groups or at parties, and giving or receiving help. Items are scored on a five-point Likert-type scale ranging from 1 to 5. The scale was designed for use with young adults (Smith & Betz, 2000); it is psychometrically sound and widely used in counseling psychology, career assessment and development, and cross-cultural adjustment assessment. The internal consistency for the PSSE in the current study was 0.96. Procedures The data for this study were gathered through an online survey. Previous research revealed that assessments using the PGIS-II were invariant across two data collection methods (paper-pencil versus online questionnaire) and found no difference between the two (Weigold, Weigold, & Russell, 2013). First, an online survey was launched. Then an email explaining the nature of the study, along with a link to the survey, was sent to all of the international graduate students studying in a large Midwestern university, with the help of International Student Services. The survey was approximately 10 min in length and no incentive was offered for participation. After a period of data collection, the survey was closed and the data were screened for missing information. A total of 271 participants consented to take part in the study, 69% of whom (n = 188) completed the survey. As for the unfinished surveys, more than half of the participants decided to quit taking the survey during or after completing the demographic information questionnaire (n = 44), and the remaining participants (n = 39) quit without completing the last instrument on social self-efficacy. Thus, the final data set was drawn from 223 participants who completed responses to the demographic questions and the PGIS-II, and 188 participants who completed the entire survey (with a small portion of responses missing from the social self-efficacy scale). Thus, 223 of the survey responses were utilized for confirmatory factor analysis (CFA), and 188 for all of the other analyses. The small portion of missing responses for the last questionnaire (fewer than 1% of the total responses) was handled by averaging the available responses for the questionnaire, as suggested by Smith and Betz (2000). Results The demographic information of the participants is summarized in Table 1. The descriptive statistics (means and standard deviations) and zero-order correlations for the measures are presented in Table 2. Consistent with our hypothesis, there was a significant positive correlation between PGI and social self-efficacy (r = 0.45, p ≪ 0.001). Significant positive correlations were found among and between all the subscales of the PGIS-II and social self-efficacy. The associations between social self-efficacy and planfulness, and social self-efficacy and intentional behavior, were equally strong (r = 0.40, p ≪ 0.001). Evaluation of the factorial structure of PGIS-II First, CFA were conducted to examine whether multidimensional factorial structure of the PGIS-II would be supported in our sample. The following indices were used to examine model fit: the relative chi-square (χ2/df), the comparative fit index (CFI), the root-mean-square error of approximation (RMSEA) with 90% confidence intervals, and the standardized root-mean-square residual (SRMR). The following criteria were used to evaluate the adequacy of model fit: a χ2/df ≪ 5 (Marsh & Hocevar, 1985), a CFI value equal to or higher than 0.90, a RMSEA value of less than 0.08 (Browne & Cudeck, 1993), and a SRMR lower than 0.08 (Hu & Bentler, 1999). Three alternative models suggested in previous research studies were compared. The model fit indices for the one-factor (Model 1), first-order four-factor (Model 2), and second-order four-factor (Model 3) solutions are presented in Table 3. In Model 1, all 16 of the PGIS-II items loaded on a general PGI factor. In Model 2, the four sub-dimensions of the PGIS-II served as four separate and correlated factors. Lastly, Model 3 had a hierarchical structure, in which all of the four sub-dimensions of the PGIS-II loaded onto a second-order PGI factor. The one-factor model (Model 1), in which the single PGI factor loaded on all 16 items, yielded a poor model fit: χ2/df = 6.43, CFI = 0.76, RMSEA = 0.16 (90% CI = 0.15–0.17), and SRMR = 0.08. An examination of the modification indices 91

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

Table 1 Demographic Information. Demographic Variable Gender. Male Female Length of residence in the US Less than a year 1–2 years 2–3 years 3–4 years More than 4 years Education level Master's program Doctoral program Ethnicity European Asian Middle Eastern African Latino Other Marital status Single Dating for more than 6 months Married Other (Divorced, Widowed and etc.) English language proficiency Poor Fair Satisfactory Good Excellent

n

%

100 88

53 47

37 34 26 34 57

20 18 14 18 30

80 108

43 57

20 112 17 4 25 10

11 60 9 2 13 5

104 24 58 2

55 13 31 1

6 25 46 63 48

3 13 25 34 25

N = 188. Table 2 Descriptive Statistics and Zero-Order Correlations Among the PGIS-II Scales and Social Self-efficacy Measures

1

2

3

4

5

6

1. PGI. 2. Readiness for Change 3. Planfulness 4. Using resources 5. Intentional Behavior 6. Social Self-efficacy Mean SD

– 0.88*** 0.87*** 0.75*** 0.84*** 0.45*** 3.63 0.76

– 0.80*** 0.50*** 0.63*** 0.33*** 3.68 0.88

– 0.44*** 0.69*** 0.40*** 3.58 0.91

– 0.50*** 0.38*** 3.33 1.02

– 0.40*** 3.94 0.85

– 3.11 0.75

***p ≪ 0.001. Table 3 Goodness of Fit Indices of Alternative Models. Models

χ2 (n = 223)

Df

χ2/df

CFI

RMSEA (90% CI)

Model 1. 1st covariance1 2nd covariance2 Model 2 (with two covariances) Model 3 (with two covariances) Model 3 vs. Model 2

668.27 536.18 487.4 304.23 316.36

104 103 102 96 98 2

6.43 5.21 4.78 3.17 3.23

0.761 0.817 0.837 0.912 0.908

0.156 0.137 0.130 0.099 0.100

(0.145 (0.126 (0.119 (0.086 (0.088

− − − − −

0.167) 0.149) 0.142) 0.111) 0.112)

SRMR

Δχ2

0.079 0.067 0.065 0.050 0.053 12.13

Note: 1 Covariance between V14-V6; 2 Covariance between V16-V11; Model 1 refers to the one-factor model; Model 2 refers to the first-order four-factor model; Model 3 refers to the second-order four-factor model.

revealed the addition of two significantly correlated error covariances to improve model fit. A scrutiny of these items shows that they were phrased similarly. The first error covariance involved items 6 and 14 on the Using Resources subscale, where both items measured asking support for self-development. The first error covariance was also commonly 92

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

found in previous research studies using the PGIS-II (i.e. Freitas et al., 2017; Yakunina, Weigold, Weigold et al., 2013; Weigold et al., 2014). The second error covariance involved items 11 and 16 on the Readiness for Change subscale, where both items assessed selfawareness to make specific changes in oneself. Given item content similarity, items 11 and 16 were allowed to covary to improve model fit. The final results of Model 1, with the addition of the two error covariances, still suggest a lack of good fit: χ2/df = 4.78, CFI = 0.84, RMSEA = 0.13 (90% CI = 0.12–0.14). The first-order four-factor solution (Model 2) offered a better fit, although, due to the RMSEA value, it still was not good. With the addition of the two error covariances, the fit indices of Model 2 were: χ2/df = 3.17, CFI = 0.91 RMSEA = 0.10 (90% CI = 0.09–0.11), and SRMR = 0.05. As the final alternative model, the second-order four-factor solution (Model 3) was assessed, in which the four dimensions served as four factors and PGI was the second-order factor. The goodness of fit indices for the second-order four-factor solution (Model 3) with the two-error term covariance added (χ2/df = 3.23, CFI = 0.91, RMSEA = 0.10 (90% CI = 0.09–0.11), and SRMR = 0.05) were quite similar to the fit indices of the first-order four-factor solution (Model 2). A chi-square difference test was conducted between the first- and second-order four-factor models (Model 2 and Model 3), and the results revealed a significant difference: χ2 (2, n = 223) = 12.13, p ≪ 0.001. This finding is consistent with prior research and provided further evidence for the proposed multidimensional first-order four-factor structure of the construct (Model 2). However, when considering model parsimony, the second-order four-factor model (Model 3) is more parsimonious (Chen, Sousa, & West, 2005). Thus, we also used an overall PGI score (second-order four-factor model) when evaluating the construct in subsequent analyses. Hierarchical regression analyses Hierarchical multiple regression analysis was conducted to examine whether social self-efficacy predicts PGI above and beyond length of residence in the U.S. and English language proficiency. Researchers have identified English language proficiency and length of residence in the U.S. as two of the variables related to cross-cultural adjustment (Wilton & Constantine, 2003; Yeh & Inose, 2003; Zhang & Goodson, 2011). Since PGI was shown to be an indicator of better cross-cultural adjustment, prior research has assessed English language proficiency and length of residence in the U.S. along with PGI (Yakunina, Weigold, Weigold et al., 2013). A series of hierarchical multiple regression analyses with three separate steps were conducted, in which each of the four primary factors of PGI, as well as the overall PGI, served as the dependent variable. Step 1 included the predictor length of residence in the U.S, followed by the English language proficiency variable in Step 2, and Step 3 included the social self-efficacy scores. We were interested in examining whether social self-efficacy would lead to a significantly greater amount of variance in predicting PGI and each of the four factors of PGI above and beyond demographic and linguistic factors. The results of the hierarchical regression analysis showed that length of residence in the U.S. was a significant predictor only for Using Resources. Step 2, with English language proficiency, significantly predicted the overall PGI, and three factors of PGI. It is worth noting that English language proficiency failed to serve as a significant predictor only for the Using Resources factor. Moreover, the social self-efficacy scores included in the regression models in the last step significantly predicted overall PGI and each of the four factors. Overall, the regression model accounted for 22% of the variance in predicting the PGI (R = 0.47, F(3,184) = 16.925, p ≪ 0.001), 13% of the variance in predicting Readiness for Change (R = 0.37, F(3,184) = 9.413, p ≪ 0.001), 17% of the variances for Planfulness (R = 0.41, F(3,184) = 12.359, p ≪ 0.001) and Using Resources (R = 0.41, F(3,184) = 12.242, p ≪ 0.001), and 19% of the variance for Intentional Behavior (R = 0.43, F(3,184) = 13.953, p ≪ 0.001). The results support our hypothesis that social self-efficacy predicted PGI, as well as each factor of PGI, even after taking into account demographic and linguistic factors(Table 4) Discussion For this study, we assessed the factorial structure of the PGIS-II, especially in light of calls by researchers to further evaluate the structure of the PGIS-II (Freitas et al., 2016; Yakunina, Weigold, Weigold et al., 2013). In addition, we examined whether efficacious perceptions of social skills (i.e., social self-efficacy) predicted PGI for international students. Evaluation of the factorial structure of the PGIS-II Our findings aligned with our expectations and were consistent with the initial development study of the PGIS-II (Robitschek et al., 2012). The results of the confirmatory factor analysis provided evidence for the stability and multifaceted structure of the scale in an ethnically diverse sample of students. As found in the majority of previous research studies (e.g., Robitschek et al., 2012; Weigold et al., 2014; Yakunina, Weigold, Weigold et al., 2013), our results provided further evidence to the four-factor structure of PGIS-II. The Cronbach’s alpha of the total scale and the subscales, as well as their mean scores and standard deviations, were similar to the descriptive statistics reported in the initial study (see Robitschek et al., 2012). When evaluating the factorial structure of PGIS-II, the three proposed factorial models for the instrument were tested. Our fit indices for each model were very similar to the results of the previous studies that examined the same structures. Consistent with prior empirical research, the one-factor model yielded a poor model fit, while the fit of the first-order and second-order four-factor models with correlated errors were both adequate and quite similar (e.g., Freitas et al., 2017; Weigold et al., 2014; Yakunina, Weigold, Weigold et al., 2013). Thus, we evaluated how each of the four factors of PGI served as a unique predictor of the independent variable, social self-efficacy. Also, we calculated a total PGI score to assess its relation with social self-efficacy, since theoretical justification and model parsimony gave plausibility to evaluate PGI as a general construct composed of four dimensions. The fact, however, remains that examining each of the four dimensions of PGI in relation to other constructs could provide 93

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

Table 4 Hierarchical Multiple Regression Analyses for Predicting PGI, and Each Four Factor of PGI. B.

SE.

β

R2

PGI Total Length of residence in the US English Language Proficiency Social Self-efficacy

0.05 0.17 0.42

0.04 0.05 0.07

10 0.24** 0.41***

0.22

Readiness for Change Length of residence in the US English Language Proficiency Social Self-efficacy

0.06 0.20 0.32

0.04 0.06 0.09

0.09 0.24** 0.28***

0.13

Planfulness Length of residence in the US English Language Proficiency Social Self-efficacy

0.05 0.18 0.44

0.04 0.06 0.09

0.09 0.21** 0.37***

0.17

Using Resources Length of residence in the US English Language Proficiency Social Self-efficacy

0.12 0.08 0.51

0.05 0.07 0.10

0.19* 0.09 0.38***

0.17

Intentional Behavior Length of residence in the US English Language Proficiency Social Self-efficacy

−0.03 0.22 0.39

0.04 0.06 0.08

−0.05 0.28*** 0.35***

0.19

N = 188. * p ≪ 0.05. ** p ≪ 0.01. *** p ≪ 0.001.

meaningful information such as revealing the unique roles of each dimension. Moreover, we added two sets of residual terms to improve model fit. Items 6 (“I ask for help when I try to change myself”) and 14 (“I actively seek out help when I try to change myself”), as well as items 11 (“I know when I need to make a specific change in myself”) and 16 (“I know when it’s time to change specific things about myself”) were allowed to co-vary in order to improve the model fit. A closer examination of these residual terms revealed that the first highly correlated items (6 and 14) belong to the Using Resources subscale, and the second residual term (items 11 and 16) pertain to the Readiness for Change subscale. Replication studies among ethnically different samples produced similar results with regard to the inclusion of residual terms in order to obtain a satisfactory fit (e.g., Freitas et al., 2017; Weigold et al., 2014; Yakunina, Weigold, Weigold et al., 2013). The two error terms reported in the present study were likely due to the similar content of the items. In our results, the Using Resources subscale had the lowest mean score and the largest variance among the subscales. This finding was consistent with findings reported in the initial study on the PGIS-II, conducted with various samples of European Americans (Robitschek et al., 2012). Robitschek et al. (2012) suggested that the Using Resources dimension, which refers to the utilization of external resources for bringing about the growth process, might be more adaptive in samples drawn from collectivist populations (i.e., Asians and Mexicans), as they tend to value interdependence and supporting each other. However, our results, like those of a similar study conducted with a predominantly Asian international student sample (Yakunina, Weigold, Weigold et al., 2013), did not provide support for this assumption. A possible reason for this might be related to the findings of Taylor et al. (2004), that Asians and Asian Americans, in fact, tend to engage in less social help-seeking than European Americans. Considering the important fact that the adverse impacts of acculturative stress were significantly smaller for international students who showed high levels of seeking out support (Yakunina, Weigold, Weigold et al., 2013), future research is needed to understand why international students who come from collectivist cultures tend to rely less on external support resources and engage in less (social) help-seeking behaviors. Consistent with previous research studies, the strongest correlation among the subscales of the PGIS-II was found between the Readiness for Change and the Planfulness subscales (r = 0.80, p ≪ 0.001). Robitschek et al. (2012) pointed out for an evaluation of these subscales to determine whether they were distinct from each other or could be captured under a common factor due to their strong correlation. The magnitude of correlation between the two subscales is quite large, which makes sense because they both belong to the cognitive dimensions of PGI (Freitas et al., 2017). However, their strong relationship and the residual terms between items in the Readiness for Change and Planfulness subscales, identified in prior research (i.e., Weigold et al., 2014), may also warrant further re-evaluation of these two subscales. Social self-efficacy as a predictor of PGI and its factors Previous research suggests that language proficiency and length of stay in the U.S. are indicators of cross-cultural adjustment (Wilton & Constantine, 2003; Xu, 1991), which is positively linked to PGI. Thus, we controlled for the effect of these potential adjustment factors when examining the relations between social self-efficacy and each factor of PGI. The results show that language 94

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

proficiency, but not length of stay, accounted for a significant, but small and unique, variance in total PGI and three subscales of the PGIS-II, except Using Resources. Length of stay in the U.S. accounted for a significant amount of variance in the subscale of Using Resources. The results indicate that the higher the language proficiency of respondents, the more likely they will be prepared to initiate the change process, come up with a plan to carry out throughout the change process, and actively involved in the change process. Furthermore, the longer the respondents have lived in the U.S., the more likely they will seek out external resources to be able to develop and grow as a person. This finding makes sense, as suggested by Wilton and Constantine (2003), in that students with greater length of stay in the U.S. have access to and utilization of greater resources than those who recently move to the U.S. Comparisons among the hierarchical regressions with the four subscales of PGIS-II as dependent variables show that when there was a one-unit change in social self-efficacy, corresponding changes were smallest for the subscale of Readiness for Change, compared to the other three subscales and the total PGI score. Readiness for Change is one of the subscales for the cognitive aspects of PGIS-II; it measures one’s preparedness for making specific change in one’s self, which is one aspect of a person’s mindset. Thus, it is plausible to conclude that a person’s mindset is more difficult to change than her or his behaviors. The finding of a positive correlation between language proficiency and social self-efficacy is not surprising, considering the fact that language proficiency is often needed for communication and social interaction (Lin & Betz, 2009). Importantly, the results from the regression analyses demonstrate that social self-efficacy provides unique prediction of PGI, even after perceived language skills are taken into account. Students with high levels of social self-efficacy are likely those who also invest more conscious effort in their personal development and growth, regardless of their English proficiency. Prior studies have shown that social self-efficacy is positively associated with secure attachments to others (Wei, Russell, & Zakalik, 2005) and to the formation of supportive friendships in a new environment (Meng, Huang, Hou, & Fan, 2015). Furthermore, it is negatively associated with depression and loneliness (Wei et al., 2005), and acculturative stress (Lin & Betz, 2009). Thus, the belief or feelings that one is socially competent may serve as a personal strength or asset to alleviate the challenges associated with adapting to a new environment (e.g., loneliness and adjustment problems), and strengthen the social skills needed to function appropriately in the host country (e.g., making friends). All of these, in turn, might facilitate cultural learning, cross-cultural adaptability, and the self-change process. Aside from cross-cultural adaptability and the self-change process, our findings support the view that both cognitive and behavioral processes play a role in cross-cultural adaptation. The relation between PGI and social self-efficacy was not unexpected, given that PGI contains “the cognitive components of self-efficacy, including beliefs, attitudes, and values that support personal growth” (Robitschek, 1998 p. 184). As social self-efficacy is one of the domains of the general self-efficacy construct, it may be reasonable to expect a stronger positive relation between social self-efficacy and the cognitive aspects of PGI. However, the correlation coefficients between social self-efficacy and the subscales of PGI were all of moderate magnitude. This finding may stem from the instrument used to assess social self-efficacy, which tapped the behavioral aspects of social functioning more than the cognitive aspects (Grieve, Witteveen, Tolan, & Jacobson, 2014). The finding that social self-efficacy beliefs and PGI are linked has implications for enhancing high-impact integration practices for the cross-cultural transitioning experiences of international students. The results of the study suggest that promoting growth opportunities for international students is possible by increasing their social skills, social competence, and social self-efficacy. To boost self-development and growth, personnel and advisors working with international students should enhance students’ social skills to facilitate their interactions with people in the host nation, including those of their own nationality as well as those of other nationalities. Based on the findings of prior research, several ways to nurture social skills, self-efficacy, and connectedness are: to increase social support resources on campus, to strengthen multicultural competence for all university personnel and students, to offer social skills training, to create a friendly and diverse campus climate, and to encourage home sharing among local and international students (Hendrickson, Rosen, & Aune, 2011; Moores & Popadiuk, 2011). Limitations and direction for future research Several limitations of this study should be acknowledged. First, due to the cross-sectional nature of the data, causal conclusions cannot be drawn from the results about the direction of influence between social self-efficacy and PGI. Future studies with longitudinal or experimental designs would help address questions regarding the directionality of influence between the study variables. Also, prospective studies that would follow students before and during their cross-cultural transition could clarify how social selfefficacy and PGI may influence one another across time. It is also important to acknowledge that all measures were self-reports based on subjective evaluations of perceptions and beliefs, which may produce response bias. However, it is equally essential to note that social self-efficacy and PGI are constructs that are very dependent on one’s own perceptions, feelings, beliefs, and attitudes. Conclusion In his earlier writing, Adler (1975) proposed an alternative view of the cross-cultural transitioning process, drawing attention to the positive outcomes of the transitioning experience, such as changes and learning processes that engender self-development and growth. The aim of this study was to explore international students’ PGI in relation to their perceived ability to successfully function in their daily lives. We also examined the factorial structure of the recently revised PGIS-II, an instrument used to measure PGI. Our results indicate that belief in one’s own abilities regarding social skills or social competence is linked to one’s conscious or intentional personal growth. With regard to the factorial structure of the PGIS-II, our findings provided further evidence in support of its fourfactorial structure, which has been found in the majority of the studies that have used that measure (e.g., Robitschek et al., 2012; 95

International Journal of Intercultural Relations 61 (2017) 88–96

E.M. Çankaya et al.

Yakunina, Weigold, Weigold et al., 2013; Weigold et al., 2014). References Adler, P. S. (1975). The transitional experience: An alternative view of culture shock. Journal of Humanistic Psychology, 15, 13–23. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H Freeman. Bhattacharya, A., & Mehrotra, S. (2014). Personal growth initiative scale: How does it perform in Indian youth samples? Journal of the Indian Academy of Applied Psychology, 40(1), 122–129. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen, & J. S. Long (Eds.). Testing structural equation models (pp. 136–162). Beverley Hills, CA: Sage. Chen, F. F., Sousa, K. H., & West, S. G. (2005). Teacher's corner: Testing measurement invariance of second-order factor models. Structural Equation Modeling, 12(3), 471–492. http://dx.doi.org/10.1207/s15328007sem1203_7. Freitas, C. P. P., Damásio, B. F., Tobo, P. R., Kamei, H. H., & Koller, S. H. (2016). A systematic review about personal growth initiative. Anales de Psicología, 32, 770–782. http://dx.doi.org/10.6018/analesps.32.3.219101. Freitas, C. P. P., Damásio, B. F., Kamei, H. H., Tobo, P. R., Koller, S. H., & Robitschek, C. (2017). Personal growth initiative scale-II: Adaptation and psychometric properties of the Brazilian version. Paidéia (in press). Grieve, R., Witteveen, K., Tolan, G. A., & Jacobson, B. (2014). Development and validation of a measure of cognitive and behavioral social self-efficacy. Personality and Individual Differences, 59, 71–76. http://dx.doi.org/10.1016/j.paid.2013.11.008. Harrison, J. K., Chadwick, M., & Scales, M. (1996). The relationship between cross-cultural adjustment and the personality variables of self-efficacy and self-monitoring. International Journal of Intercultural Relations, 20(2), 167–185. Hendrickson, B., Rosen, D., & Aune, R. K. (2011). An analysis of friendship networks, social connectedness, homesickness, and satisfaction levels of international students. International Journal of Intercultural Relations, 35(3), 281–295. http://dx.doi.org/10.1016/j.ijintrel.2010.08.001. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. http://dx.doi.org/10.1080/10705519909540118. Lillyman, S., & Bennett, C. (2014). Providing a positive learning experience for international students studying at UK universities: A literature review. Journal of Research in International Education, 13(1), 63–75. http://dx.doi.org/10.1177/1475240914529859. Lin, S., & Betz, N. E. (2009). Factors related to the social self-efficacy of Chinese international students. The Counseling Psychologist, 37(3), 451–471. http://dx.doi.org/ 10.1177/0011000009332474. Marsh, H. W., & Hocevar, D. (1985). Application of confirmatory factor analysis to the study of self-concept: First- and higher-order factor models and their invariance across groups. Psychological Bulletin, 97, 562–582. Meng, H., Huang, P., Hou, N., & Fan, J. (2015). Social self-efficacy predicts Chinese college students’ first-year transition: A four-wave longitudinal investigation. Journal of Career Assessment, 23, 410–426. Moores, L., & Popadiuk, N. (2011). Positive aspects of international student transitions: A qualitative inquiry. Journal of College Student Development, 52(3), 291–306. Ogunyemi, A. O., & Mabekoje, S. O. (2007). Self-efficacy, risk-taking behavior and mental health as predictors of personal growth initiative among university undergraduates. Electronic Journal of Research in Educational Psychology, 5(2), 349–362. Robitschek, C., & Kashubeck, S. (1999). A structural model of parental alcoholism, family functioning, and psychological health: The mediating effects of hardiness and personal growth orientation. Journal of Counseling Psychology, 46, 159–172. Robitschek, C., & Keyes, C. L. M. (2009). Keyes’s model of mental health with personal growth initiative as a parsimonious predictor. Journal of Counseling Psychology, 56(2), 321–329. http://dx.doi.org/10.1037/a0013954. Robitschek, C., Ashton, M. W., Spering, C. C., Geiger, N., Byers, D., Schotts, G. C., & Thoen, M. A. (2012). Development and psychometric evaluation of the Personal Growth Initiative Scale-II. Journal of Counseling Psychology, 59(2), 274–287. http://dx.doi.org/10.1037/a0027310. Robitschek, C. (1998). Personal growth initiative: The construct and its measure. Measurement and Evaluation in Counseling and Development, 30(4), 183–198. Robitschek, C. (1999). Further validation of the personal growth initiative scale. Measurement and Evaluation in Counseling and Development, 31(4), 197–210. Sharma, H. L., & Rani, R. (2013). Relationship of personal growth initiative with self-efficacy among university postgraduate students. Journal of Education and Practice, 4(16), 125–135. Smith, H. M., & Betz, N. E. (2000). Development and validation of a scale of perceived social self-efficacy. Journal of Career Assessment, 8(3), 283–301. Taylor, S. E., Sherman, D. K., Kim, H. S., Jarcho, J., Takagi, K., & Dunagan, M. S. (2004). Culture and social support: Who seeks it and why? Journal of Personality and Social Psychology, 87, 354–362. Tedeschi, R. G., & Calhoun, L. G. (2004). Target article: Posttraumatic growth: Conceptual foundations and empirical evidence. Psychological Inquiry, 15(1), 1–18. http://dx.doi.org/10.1207/s15327965pli1501_01. Wei, M., Russell, D. W., & Zakalik, R. A. (2005). Adult attachment, social self-efficacy, comfort with self-disclosure, loneliness, and subsequent depression for freshmen college students: A longitudinal design. Journal of Counseling Psychology, 52, 602–614. Weigold, A., Weigold, I. K., & Russell, E. J. (2013). Examination of the equivalence of self-report survey-based paper-and-pencil and internet data collection methods. Psychological Methods, 18(1), 53–70. http://dx.doi.org/10.1037/a0031607. Weigold, I. K., Weigold, A., Russell, E. J., & Drakeford, N. M. (2014). Examination of the psychometric properties of the Personal Growth Initiative Scale-II in African American college students. Assessment, 21(6), 754–764. http://dx.doi.org/10.1177/1073191114524019. Williams, T. R. (2005). Exploring the impact of study abroad on students’ intercultural communication skills: Adaptability and sensitivity. Journal of Studies in International Education, 9, 356–371. Wilton, L., & Constantine, M. G. (2003). Length of residence, cultural adjustment difficulties, and psychological distress symptoms in Asian and Latin American international college students. Journal of College Counseling, 6(2), 177–186. http://dx.doi.org/10.1002/j. 2161-1882.2003. tb00238.x. Xu, M. (1991). The impact of English-language proficiency on international graduate studentsʼ perceived academic difficulty. Research in Higher Education, 32, 557–570. Yakunina, E. S., Weigold, I. K., & Weigold, A. (2013). Personal growth initiative: Relations with acculturative stress and international student adjustment. International Perspectives in Psychology: Research, Practice, Consultation, 2(1), 62–71. http://dx.doi.org/10.1037/a0030888. Yakunina, E. S., Weigold, I. K., Weigold, A., Hercegovac, S., & Elsayed, N. (2013). International students’ personal and multicultural strengths: Reducing acculturative stress and promoting adjustment. Journal of Counseling and Development, 91(2), 216–223. http://dx.doi.org/10.1002/j.1556-6676.2013.00088.x. Yalçın, İ., & Malkoç, A. (2013). Adaptation of personal growth initiative scale- II to Turkish and investigation of psychometric properties. The Journal of Psychiatry and Neurological Sciences, 26(3), 258–266. Yang, H., & Chang, E. C. (2014). Examining the structure, reliability, and validity of the Chinese personal growth initiative scale–II: Evidence for the importance of intentional self-change among Chinese. Journal of Personality Assessment, 96, 559–566. Yeh, C. J., & Inose, M. (2003). International students’ reported English fluency, social support satisfaction, and social connectedness as predictors of acculturative stress. Counseling Psychology Quarterly, 16(1), 15–28. Zhang, J., & Goodson, P. (2011). Predictors of international students’ psychosocial adjustment to life in the United States: A systematic review. International Journal of Intercultural Relations, 35(2), 139–162. http://dx.doi.org/10.1016/j.ijintrel.2010.11.01.

96