- Email: [email protected]
Antarctic stations as workplaces: Adjustment of winter-over crew members
Polar Science 22 (2019) 100484
Contents lists available at ScienceDirect
Polar Science journal homepage: http://www.elsevier.com/locate/polar
Antarctic stations as workplaces: Adjustment of winter-over crew members Cyril Jaksic, Gary Steel *, Emma Stewart, Kevin Moore Department of Tourism, Sport & Society Faculty of Environment, Society & Design P O Box 85084 Lincoln University, Lincoln, 7647, Christchurch, New Zealand
A R T I C L E I N F O
A B S T R A C T
Keywords: Antarctica Isolated and confined environments Extreme environment Psychological Adaptation Person-environment fit Polar psychology
The living conditions at Antarctic stations can be challenging for support personnel. It has been; suggested that the experience of isolation and confinement can contribute to the emergence of the; winter-over syndrome. The present study adopts a Person-Environment fit approach to investigate; individual adjustment to the social constraints of an Isolated and Confined Environment (ICE). The; study gathered monthly data from 14 partici pants from five different stations, run by different National Antarctic Programmes. Results revealed that a lack of privacy generated by the confinement is associated with sleep disturbance. In addition, a high level of loneliness, experienced as a result of the; isolation, is positively related to cognitive impairment and negatively related to job satisfaction and; positive/negative mood ratio. The results further suggest that loneliness can be predicted by a predeployment; measure of need for affiliation, as well as levels of the personality traits of agreeableness; and extraversion.
1. Introduction Human activity in Antarctica has typically been predominantly focused on scientific research. Most Antarctic stations, which offer the facilities needed for such research, are generally operated by National Antarctic Programmes (NAP). Each NAP is responsible for selecting the science and support personnel that will be deployed to their station. The ability to adjust in a positive way to such environments varies from individual to individual (Peri et al., 2000; Roy and Deb, 1999). It is, therefore, important to recruit people who will best fit with critical features of this environment in order to increase the likelihood of pos itive outcomes, both professional and personal. The polar environment presents its own set of unique features in regards to human psychology. Those going to Antarctic stations have to deal with the remoteness of the continent. This remoteness places con straints on the number and types of interactions that can occur with the world north of 60� S with sometimes limited phone or internet access. Such social isolation has been noted as being amongst the major stressors in Antarctica (McCormick et al., 1985). Separation from family and friends was also the most frequently mentioned source of stress for two expeditions that spent 17 days in complete isolation in the Arctic (Bishop et al., 2001). It was also ranked as the main stressor by different Australian crews wintering-over in Antarctica between 1980 and 1982 (Godwin, 1986). Strange and Klein (1973) found that almost all anxious feelings reported by U.S. personnel over the winter were related to
events occurring back home. It is not helping that perceived peer sup port degrades over the winter-over (Nicolas et al., 2016). It is important to note that while the term ‘isolation’ might refer to a physical description (being physically isolated), it only becomes a problem when the isolation from the usual social network is negatively experienced and leads one to feel lonely. It is acknowledged that isola tion does not systematically lead to feeling lonely, nor that it is a necessary precondition for it, but rather it can facilitate the emergence of such a feeling. The distinction between isolation and loneliness is important because it is possible for two individuals to be in the same physically isolated environment but not feel equal levels of loneliness; for some, it may even be a motivating factor for working in such a re gion. The present paper will not consider the objective isolation but, instead, the negative experience of a perceived isolation, in other words; loneliness. In addition to loneliness, confinement can also take its toll on Ant arctic personnel. They must cope with having to share living spaces with workmates, scientists, and VIP visitors, usually in very close quarters. This includes sharing facilities such as bathrooms, dining halls, and leisure rooms. During the busiest season (summer), personnel usually share bedrooms. There are often limited opportunities to venture beyond the immediate confines of the station, especially in winter due to extreme weather or other safety considerations. The effects that these psychological challenges may generate are even more prevalent for those who spend an entire winter in an Antarctic station; so-called
* Corresponding author. E-mail address: [email protected] (G. Steel). https://doi.org/10.1016/j.polar.2019.100484 Received 29 April 2019; Received in revised form 26 September 2019; Accepted 16 October 2019 Available online 18 October 2019 1873-9652/© 2019 Elsevier B.V. and NIPR. All rights reserved.
C. Jaksic et al.
Polar Science 22 (2019) 100484
winter-over personnel. This confinement may lead to experiencing a lack of privacy. In one study, perceived lack of privacy has been reported as the second most important source of stress by Australian winter-overs (Godwin, 1986). Cravalho (1996) mentions the case of a woman who complained that even when she wanted to get away from McMurdo station, it had to be done with someone else. As noted by Binsted et al. (2010), this concern over privacy also applies to the limited auditory privacy due to the layout of polar stations. Finally, as one winter-over reported, in an Antarctica station “tu entends tout. En fait tu n’as aucune intimit� e” [you hear everything. Actually, you have no privacy] (Solignac, 2004, p. 130). It has been suggested (Palinkas, 1992, 2002) that both loneliness and confinement could be partially responsible for commonly observed symptoms amongst winter-overs, clustered together under the name of the winter-over syndrome. The winter-over syndrome is characterised by negative mood (depression and hostility), sleep disturbance and impaired cognition (Strange and Klein, 1973). The notion that isolation and confinement, and its potentially concomitant loneliness and lack of privacy, can contribute to the emergence of the winter-over syndrome is supported by studies conducted outside of Antarctica. For instance, it has been shown that social isolation was related to sleep disturbance (Cacioppo et al., 2000; Hawkley et al., 2010) and cognitive impairment (e.g., memory, attention, information processing speed) (Ellwardt et al., 2013). Further, indirect evidence for the link between confinement and the same outcome variables is provided by research on experiences of crowding. It has also been found that crowding is related to negative mood (Nagar and Pandey, 1987; Smith and Haythorn, 1972; Zee dyk-Ryan and Smith, 1983), sleep disturbance (Rona et al., 1998) and cognitive impairment (Nagar and Pandey, 1987). Both crowding and issues of privacy arise when human density is perceived as higher than desired. In addition, feeling crowded conceptually entails that one does not have adequate privacy. Given these findings, it seems plausible that both isolation and confinement might be contributing to the emergence of the winter-over syndrome. Isolation and confinement can also adversely affect a whole team, and even the NAP itself. For instance, at least one field accident has been attributed to the distraction caused by the first mail delivery from an Antarctica expeditioner’s family (Taylor and McCormick, 1987). Pal inkas (2002) reported the case of two men who had to be sent back home before the end of their mission when preoccupations with their family prevented them from efficiently carrying out their assignment. For all NAP, sending individuals to the Ice and keeping the stations running is expensive. Replacing an employee can be costly and logistically difficult. Therefore, adjusting to this unusual environment is crucial for the in dividual’s well-being, the whole team, and the NAP. In order to cope with the lack of privacy, Jenkins and Palmer (2003) have proposed that social withdrawal could be an effective regulation strategy. It is assumed that, to a certain extent, the negative conse quences of a lack of privacy can be thwarted by some strategies, either cognitive (e.g., mentally shutting out other individuals) or behavioural (e.g., physically withdrawing in one’s room or outside). As an example, Cravalho (1996) reported that a crew member started working nights to limit social interactions. The lack of privacy must, therefore, be considered in conjunction with the strategies put in place to preserve one’s privacy. To investigate adjustment to an Antarctic station in this study, a specific theoretical approach was adopted: Person-Environment fit theory (henceforth referred to as P-E fit). This theory is based on the assumed interdependence of an individual’s and an environment’s characteristics. It posits that, for relevant environmental features, the match with one individual’s corresponding characteristics can predict general adjustment to that environment. In Antarctica, therefore, the extent to which one adjusts to the isolation and confinement should predict the extent of one’s winter-over syndrome, job performance and job satisfaction.
Because the major issue with isolation is the potential frustration of one’s need for directly socialising with one’s normal social network, it is expected that needs associated with social relationships are the most relevant predictors of one’s adjustment to isolation. A number of such needs have been identified (see, e.g., Ryan and Deci, 2000). We consider here the need for affiliation and the need for intimacy. The need for affiliation has been defined as “establishing relationships €nbrodt and Gersten to rather unfamiliar people and acquaintances” (Scho berg, 2012, p. 4) and as “the urge to form connections and make contact with other people” (Vaughan and Hogg, 2014, p. 503). The need for in timacy is concerned with “being close to others, having positive profound interactions, and practicing self-disclosure and warm mutual exchange” €nbrodt and Gerstenberg, 2012, p. 4) and “experiencing a warm, (Scho close, and communicative exchange with another person” (McAdams, 1980, p. 413). The critical differences between the two concepts are those of depth and closeness. The need for intimacy pushes one to seek re lationships that manifest these qualities, whereas the need for affiliation simply motivates one to make social contacts, however superficial. Taken together, though, it is expected that higher levels of need for affiliation and need for intimacy will put a greater demand on what can be provided in the social environment, thus increasing the likelihood of loneliness in an Antarctic setting. The degree of confinement occasionally experienced in an Antarctic station can interact with the need for privacy and the need for intimacy. Confinement, that potentially leads to a lack of privacy, will present greater challenges to an individual with a high need for privacy. Simi larly, a high need for intimacy requires that individuals have sufficient privacy to bond and develop a close relationship with one or more other people. As a result, confinement will be more challenging for individuals with high need for intimacy. However, the relationship between one’s needs and the extent to which they manifest themselves is not straightforward. The channelling hypothesis (Winter et al., 1998) suggests that, while the needs determine our goals, personality traits can affect the means used to reach them. For instance, two individuals with a high need for intimacy but differing levels of extraversion will not necessarily behave the same way in the same context. One who is extraverted might feel comfortable making many relationships in the hope that a few will provide intimacy. In contrast, an introvert might choose to spend time with only a few people but will put extra effort into creating intimacy. With the focus of this study on social needs predicting social adjustment, two personality traits from the Big Five (Goldberg, 1981) have been identified as potentially influential channels, namely agreeableness and extraversion. Those two personality traits have been systematically found to correlate with social variables (e.g., social support, interpersonal competence) (Judge et al., 1999; Lopes et al., 2004; Lopes et al., 2005). Moreover, they are the only two traits that are fundamentally defined by the way we interact with others (extraversion; sociability, assertiveness, talkativeness/agree ableness; altruistic, sympathetic, helpful) (Rothmann and Coetzer, 2003). On the other hand, neuroticism relates to how one deals with negative affects, conscientiousness relates to approaching and carrying out tasks and openness to experience relates to one’s imagination and curiosity. None of which specifically deals with social environments. Therefore, agreeableness and extraversion are the most relevant traits for predicting people’s responses to the unique social environment found in Antarctic stations. The present study aimed at testing a theoretical model taking a P-E fit approach and tailored to two of the most relevant Antarctic social fea tures. The first part of the model predicts a positive relationship between pre-deployment measures of need for affiliation and intimacy with loneliness. The second part predicts a negative relationship between measures of need for intimacy and privacy with adjustment with the lack of privacy. The relationships in both parts are expected to be moderated by both extraversion and agreeableness. In addition, both loneliness and the fit with privacy are expected to be related to job satisfaction, job performance, cognitive impairment, sleep disturbance and mood. 2
C. Jaksic et al.
Polar Science 22 (2019) 100484
2. Method
(e.g., talkative, reserved, inventive). The participants had to assess whether each attribute describes them using a 5-point Likert-type response format. To record the importance of the need for social contact, each partic ipant completed two subscales from the Unified Motives Scales €nbrodt and Gerstenberg, 2012): the need for affiliation and need (Scho for intimacy scales. Respondents were asked to evaluate 10 affiliation and 10 intimacy items using a 5-point Likert-type response format. In order to measure the need for privacy, the Preference for Solitude Scale (Burger, 1995) was used. This questionnaire proposes 12 pairs of statements (e.g., Time spent alone is often productive for me vs Time spent alone is often time wasted for me) from which respondents chose the one that described them the best. Though the Preference for Solitude Scale measures the inclination toward solitude, Long and Averill (2003) and Westin (2003) report that solitude is a type of privacy. Westin (1967, p. 31) even considers solitude as “the most complete state of privacy that individuals can achieve”. Other researchers have similarly regarded the two concepts as being part of the same construct (Stewart and Cole, 2001). Hence, it is used in this study as a proxy measure for need for privacy. A high preference for solitude should reflect a high need for privacy.
2.1. Recruitment Fifteen NAP that have at least one permanent station in Antarctica were contacted via email in 2015. These included: Argentina, Australia, Chile, China, France, Germany, India, Japan, New Zealand, Norway, Russia, South Africa, South Korea, the United Kingdom and the United States. Those which agreed to forward the advertisement for the present study to their future winter-overs were: China, Germany, New Zealand and Norway. However, because the New Zealand Programme only gave approval in early 2016, once their future winter-overs were already at the station, the data for their first months at the station are missing. The other NAP either failed to reply despite receiving reminders or simply explained they could not be involved in the study for diverse reasons. Despite the absence of response from the U.S. NAP, one of their future winter-overs personally contacted the researcher to take part in the study. This person had read an interview the first author had given about the study and that had been published online. 2.2. Sample Amongst the 57 winter-overs employed by the four collaborating NAP, a total of 14 (26%) of the winter-overs filled in the pre-winter-over survey and at least one monthly survey. Amongst those 14 participants, there was one in Great Wall Station (China), four at Neumayer Station III (Germany), three at Scott Base (New Zealand), five at Troll Station (Norway) and one at Amundsen-Scott South Pole Station (U.S.). There was a total of six females and eight males. With one male failing to report his age, the average age of the 13 remaining participants was 35.7 years old (SD ¼ 8.9), ranging from 24 to 52 years old. Out of the 14 partici pants, three reported having previously wintered-over at least once, 10 had never wintered-over previously, and one did not respond to this item. All participants took part in the present study during a 2016 winter-over. This research was approved by the Lincoln University Human Ethic Committee.
2.5. Monthly survey measures The second survey was completed online on a monthly basis by the winter-overs throughout their stay in Antarctica. It comprised questions directly related to the participants’ experiences at their respective sta tions for the previous 30 days. A repeated measures was adopted in order to test the existence of correlations between the variables of in terest throughout the year. The intention was not to investigate the ef fect of time but instead, to test that relationships exist overall. Making measures at only one point in time might reveal relationships that only exist at such a specific moment. Revealing significant relationships be tween variables measured throughout the year would strengthen the results and make it more generalizable to Antarctic stays in their entirety. Job satisfaction was measured through the Brief Index of Affective Job Satisfaction (BIAJS) (Thompson and Phua, 2012). This question naire asks one to assess, on a 5-point Likert-type scale, how strongly one agrees with four statements (e.g., Most days I am enthusiastic about my job). Sleep quality was assessed via three items that have already been used in an Antarctic context (Doll and Gunderson, 1971; Palinkas et al., 1989). For the three of them (difficulties falling asleep or staying asleep; waking up at night; and feeling tired during the day), the participant had to indicate the frequency they experienced them in the past thirty days answering on a 5-point Likert-type response format. Positive and negative moods are not simply two ends of a continuum but, rather, two independent constructs (Agho et al., 1992; Watson et al., 1988). For this reason, the same format of response as the Evaluative Space Grid (ESG) by Larsen et al. (2009) was used. This single-item measure consists of a 5 � 5 grid with the x-axis representing positive affect and the y-axis representing negative affect. Each scale proposes five options (not at all; slightly; moderately; quite a bit; extremely). The participants were asked to indicate, by choosing one square, the extent to which they felt positive and negative affect during the last 30 days. This measure allows calculation of a ratio between negative and positive affect. Because it has been suggested that negative affect has a weighting of about three times more than positive affect in its effect on one’s overall well-being (Fredrickson and Losada, 2005), the ratio calculated divides the positive affect score by three times the negative affect score and is labelled Mood Ratio. Cognitive impairment was assessed via four items covering four different cognitive functions, namely memory, attention, language, and thinking. The items were as follows: Over the last 30 days, how often have you experienced: 1) being forgetful 2) difficulties to focus 3)
2.3. Procedure While most studies take pre- and post-winter-over measures (Gav alas, 2011), such an approach would not be sensitive enough to capture the variations that occur in one’s adjustment during the course of a person’s stay. It was felt that a self-report score obtained at the end of the stay was likely to represent an overall appreciation of the experience. Furthermore, such answers could be unduly influenced by the cognitive and affective state at the time of reporting, which might not validly reflect the entire experience throughout the winter. For this reason, a more fine-grained approach has been used here. The study consisted of two different surveys. The first was a pre-winter-over survey that par ticipants had to complete at any time prior to their arrival in Antarctica. However, because the New Zealand programme agreed to collaborate one month after their winter-overs arrived in Antarctica, those three participants filled in the pre-winter-over survey while on the ice. For each participant, a code was generated and provided to them every time a survey link would be sent. This code was for them to enter in each survey. This allowed for participants to fill in questionnaires anony mously while allowing the researchers identify the data collected over time as coming from the same individual. 2.4. Pre-winter-over measures The pre-winter-over survey consisted of demographic questions and the following measures. Personality traits were measured using the Big Five Inventory (BFI) (John and Srivastava, 1999). This questionnaire presents 44 attributes 3
C. Jaksic et al.
Polar Science 22 (2019) 100484
difficulties to find your words or express what you meant 4) feeling confused. The participant answered using a 5-point Likert-type response format. One overall score was calculated by averaging the four answers. P-E fit in regards to isolation was measured by asking participants to complete an adapted version of the 6-item De Jong Gierveld Loneliness Scale (Gierveld and Van Tilburg, 2006). It presents six statements to which participants respond using a 5-point Likert-type response format. In effect, this produces a measure of ‘non-fit’ with the social environ ment, where one’s needs and desires regarding social interactions are left unsatisfied. This means that a low score on Loneliness reflects a good P-E fit. Because of this, this variable will be referred to as Loneliness from this point onwards. To measure P-E fit for privacy, the concept of privacy was first defined. Participants were then asked to indicate the extent to which they had enough privacy, using a 5-point Likert-type response format. In addition, they were asked what behaviour they used to regulate their privacy. They were also asked to report the frequency with which each privacy behaviour was used. However, because the measure of privacy regulation yields non-numerical data, it will not be further addressed in the present paper but will be presented in a separate publication, and scores of P-E fit for privacy will be taken from the single item measure. To assess one’s job performance in Antarctica, peer-evaluation has often been used. Such measures can show high degrees of agreement amongst peers (Nelson and Gunderson, 1962), and it is considered a valid measure of job performance in ICEs (Grant et al., 2007; Gunderson and Nelson, 1963; Palinkas et al., 2000) and in more mundane work places (Arvey and Murphy, 1998; Furnham and Stringfield, 1998). The participants were asked to name, based on their job performance, the individuals at their station they would choose if they had to recruit a team for a future expedition. For each nominee, participants had to indicate their level of acquaintance. In order to accommodate non-English speaking participants, both surveys were translated into Chinese and German. Both versions had been back-translated to ensure the accuracy of the translation. Backtranslation has been considered as a good approach to ensure accu racy in translation (McGorry, 2000), and though alternative methods have been deemed more accurate (Beaton et al., 2000), they present the disadvantage of being extremely time-consuming and costly.
In addition, other participants simply did not respond to this item. Therefore, the numerous missing data on that measure were removed from the analyses, which left no measure of Job Performance. All existing scales used in this study had been checked for internal validity and reliability by their respective authors and were found to meet the relevant criteria. Scales created by the authors of the present study (Sleep Disturbance and Cognitive Impairment) were found to have an internal consistency (Cronbach’s α) above .80. The descriptive sta tistics for the eleven variables can be found in Table 2 below. For repeatedly measured variables, an average score was first calculated for each participant and then, the mean and standard deviation across the 14 participants was computed. 3.2. Analysis Two distinct methods could have been used to analyse the data with regard to the hypotheses. First, averaging correlations consists of calculating, for each month, the correlations of two variables of interest. This method results in 12 correlations (one per month) that are averaged to have one value reflecting the overall relationship between the two variables across time. The second method consists of correlating aver ages. Here, for each month, the average values of two variables of in terest are calculated. This method results in 12 averages per variable (one per month) that can then be correlated across the 12 months. It has been demonstrated that averaging correlations is more reliable and less biased than correlating averages (Bittner et al., 1982; Dunlap et al., 1983). For this reason, the data in this study were analysed using the averaging correlations method. However, while averaging correlations appears to be the best option, it has been found that averaging raw correlations produces greater bias than averaging Fisher’ z-transformed correlations (Silver and Dunlap, 1987). The advantage of the latter method appears to be even greater when the sample size is small (Corey et al., 1998). For each month, the Spearman correlation between two variables of interest was calculated. The non-parametric Spearman correlations were chosen over the parametric Pearson correlations because the small sample size made it difficult to test for the normality of the data, which is Table 2 Descriptive statistics for all eleven variables.
3. Results 3.1. Data
Variables
Means
Standard deviations
Minimum
Maximum
Scale score range
Table 1 shows the number of participants who completed a survey in each of the 13 months of the study. Since the concern was not with objective time (e.g., how people feel in May) but rather with relative time (e.g., how people feel during their second month at the station), and because not every participant arrived in Antarctica at the same time, each winter-over’s timeframe was aligned to start with their respective arrival month. Some participants failed to fill in their monthly survey, therefore almost all months had missing data. Only two participants filled in a survey after 13 months of stay. Therefore, only the first 12 months of each participant’s responses were taken into consideration for analyses. Out of the 14 participants, one Chinese and one US-American participants were the only winter-overs at their respective station to take part in the study. As a result, the peernomination method to assess job performance did not occur for them.
Need for Affiliation Need for Intimacy Need for Privacy Agreeableness Extraversion Job Satisfaction Sleep Disturbance Mood Ratio Cognitive Impairment Loneliness Privacy Fit
2.91
0.36
2.20
3.40
1–5
2.76
0.69
1.50
3.90
1–5
6.71
1.82
3.00
9.00
0–12
3.85 3.20 3.99
0.52 0.48 0.56
3.11 2.38 2.89
4.89 3.88 4.97
1–5 1–5 1–5
2.54
0.86
1.12
4.14
1–5
0.73 2.15
0.35 0.72
0.41 1.04
1.67 3.50
0.07–1.67 1–5
2.20 4.17
0.50 0.57
1.29 2.82
3.35 4.86
1–5 1–5
N ¼ 14.
Table 1 Number of participants (n) who filled in their survey after each month spent in Antarctica. months n
1
2
3
4
5
6
7
8
9
10
11
12
13
7
13
10
12
12
12
12
14
12
11
10
9
2
4
C. Jaksic et al.
Polar Science 22 (2019) 100484
assumed by Pearson correlations. For each pair of variables of interest, 12 Spearman correlations (one per month) were obtained. Once the 12 Spearman correlations between two variables of interest were computed, they were converted using the Fisher’s z transformation. The average and standard deviation of those z scores were then obtained. Both the average and the standard deviation have been backtransformed to be expressed as correlation coefficients, thus making it easier to assess the effect size. This average was then tested against zero using a t-test to find out if the averaged correlation is significant. As testing of the model produced 18 individual correlations, a Bon ferroni correction was required to reduce the chance of a Type I error (false positive). As a result, the level of significance is divided by 18, and all the p-values presented below should be compared to an α ¼ .003125 for determining significance at the p < .05 level.
Table 4 Back-transformed Spearman correlations after a Fisher’s z transformation be tween Loneliness and Privacy Fit, and the relevant outcome variables (all df ¼ 11). Correlations with
Loneliness
Privacy Fit
3.2.1. Predicting P-E fit Due to the small sample size, it was not possible to test for the moderation effect of Agreeableness and Extraversion on the relationship between the three needs and the two measures of P-E fit. Hence, the two personality traits were taken as simple predictors of Loneliness and Privacy Fit. Both personality traits were expected to have a negative relationship with both Loneliness and Privacy Fit. First, the extent to which both P-E fits (Loneliness and Privacy Fit) were predicted by the pre-winter-over measures was examined. Table 3 shows the correlations between the two measures of P-E fit and their respective predictors. As expected, Loneliness positively correlated with the Need for Affiliation and negatively correlated with Agreeableness. However, Loneliness and the Need for Intimacy were not significantly correlated. Moreover, Loneliness was positively correlated with Extraversion, where a negative relationship was expected. As predicted, the Privacy Fit did negatively correlate with the Need for Intimacy. However, there was no significant correlation between Privacy Fit and Need for Privacy, Agreeableness, and Extraversion. All significant correlations showed modest associations between the variables.
rS r2S p-value (1tailed) rS r2S p-value (1tailed)
Job Satisfaction
Sleep Disturbance
Cognitive Impairment
Mood Ratio
-.487 .237 <.001
.219 .048 n.s.
.486 .236 <.001
-.468 .219 <.001
.190 .036 n.s.
.404 .163 <.001
.208 .043 n.s.
-.04 .002 n.s.
lack of privacy as being an issue (Godwin, 1986), it does not directly relate to aspects of adjustment. So, while concerns are expressed about the lack of privacy in ICEs (Binsted et al., 2010), it is possible that its consequences have less impact on adjustment than was previously hypothesised. Although this seems – at first glance – somewhat unlikely, it may simply be that winter-overs become aware of a lack of privacy only when they are asked about it. This could be interpreted as a sign of adaptation, inasmuch as the condition has become “usual” and, there fore, unremarkable, and of less influence on day-to-day living. This is supported by a recent study finding the use of withdrawing strategies to decrease over time (Sandal, van deVijver and Smith, 2018). The authors suggest that winter-overs might become emotionally dumb across time and thus, their situational awareness decreases. Alternatively, it is conceivable that privacy is of considerable importance but that the measure used in the present study failed to capture the construct accurately. While the item measuring Privacy Fit grossly defined privacy as one general concept, it might be beneficial to analyse the measure into its constituent parts. Privacy has been identi fied as having different components such as anonymity (not being identified in public) or not neighbouring (relationship with neighbours) (Marshall, 1974). It might therefore be important for a measure of pri vacy to cover the full range of dimensions that the construct encompasses. There are well-established links between loneliness and depression (Nutt et al., 2008) and between depression and sleep disturbances (van Winkel et al., 2017). Thus, we expected to see a relationship between Loneliness and Sleep Disturbance but we did not find any reliable evi dence of this. The link may not be as strong as was supposed, or its absence in our study may be due to environmental features. We know that sleep impairment can be partially caused by the disruption of the circadian cycle via the unusual light-cycle in Antarctica (Francis et al., 2008). It is possible that the effect of loneliness on sleep impairment is overshadowed by this factor, which masks the more subtle relationship. A larger number of participants and controlling for the effects of the light-cycle may bring the relationship to the fore. Placing loneliness at the core of model, rather than as simply an outcome variable, allowed it to be identified as a central factor relevant to cognitive functioning, job satisfaction and mood. This suggests that
3.2.2. P-E fits predicting outcome variables The second step of the analysis consisted of looking at the relation ship between the two P-E fit measures and outcome variables relevant for winter-overs, namely Job Satisfaction, Sleep Disturbance, Cognitive Impairment and Mood Ratio. Table 4 shows the correlations between Loneliness and the outcome measures. As predicted, Loneliness positively correlated with Cognitive Impairment and negatively with Job Satisfaction and Mood Ratio. However, no significant correlation was found between Loneliness and Sleep Disturbance. Contrary to expectation, a significant positive cor relation has been found between Privacy Fit and Sleep Disturbance. No significant correlation was found between Privacy Fit and Job Satis faction, Cognitive Impairment, and Mood Ratio. As in the previous table, all significant correlations showed modest associations between the variables. 4. Discussion The fact that Privacy Fit did not correlate as expected with any of the outcome variables could suggest that, though winter-overs report the
Table 3 Back-transformed Spearman correlations after a Fisher’s z transformation between Loneliness and Privacy Fit, and their respective predictors (all df ¼ 11). Loneliness Privacy Fit
rS r2S p-value (1-tailed) rS r2S p-value (1-tailed)
Need for Affiliation
Need for Intimacy
.252 .064 <.001
.052 .003 n.s. -.319 .102 .002
5
Need for Privacy
Agreeableness
Extraversion
-.129 .017 n.s.
-.228 .052 .002 008 <.001 n.s.
.278 .077 <.001 -.223 .050 n.s.
Polar Science 22 (2019) 100484
C. Jaksic et al.
NAP should pay particular attention to future winter-overs’ needs for social contact. The present study suggests that those who adjust well with the isolation, and will not feel lonely as a result, will present better indices of affective, cognitive, and occupational adjustment. One way to combat the sense of isolation, and possibly loneliness, at an Antarctic station has been to increase the quality and quantity of communication with home. However, a recent study has suggested that this may make things worse for winter-overs. It was pointed out that some winter-overs spend so much time on the Internet to communicate with their relatives that they isolate themselves from the rest of the crew (Solignac, 2004). This can threaten group cohesion and efficiency, and make the experience of isolation quite unpleasant. It was also seen that some winter-overs focussed so much on the news received regularly from their friends and family that they neglected their work and needed to be sent back home (Palinkas, 2002) or even had field accidents because of the distraction it represented (Taylor and McCormick, 1987). Thus, a balance must be struck between contact with home and contact within the isolated group, taking into account the individual’s needs for affiliation. Based on the present study, we would recommend that NAP pay particular attention to the candidates’ needs for social contact during the recruitment process. On a final note, the small sample size of the present study is not uncommon when investigating winter-overs. This field of research has, therefore, relied on replications for strengthening the confidence in our knowledge. When similar relationships are systematically found, it suggests that the real relationships must be strong. This is especially true if said relationships are observed even with small sample sizes. More over, results found with culturally diverse samples and at different sta tions suggest that observed effects are widespread.
Bishop, S.L., Grobler, L.C., Schjøll, O., 2001. Relationship of psychological and physiological parameters during an Arctic ski expedition. Acta Astronaut. 49 (3), 261–270. Bittner, J.A.C., Dunlap, W.P., Jones, M.B., 1982. Averaged Cross-Correlations with Differentially-Stable Variables: Fewer Subjects Required with Repeated Measures. Sage Publications Sage CA, Los Angeles, CA. Symposium conducted at the meeting of the Proceedings of the Human Factors Society Annual Meeting. Burger, J.M., 1995. Individual differences in preference for solitude. J. Res. Personal. 29 (1), 85–108. Cacioppo, J.T., Ernst, J.M., Burleson, M.H., McClintock, M.K., Malarkey, W.B., Hawkley, L.C., et al., 2000. Lonely traits and concomitant physiological processes: the MacArthur social neuroscience studies. Int. J. Psychophysiol. 35 (2), 143–154. Corey, D.M., Dunlap, W.P., Burke, M.J., 1998. Averaging correlations: expected values and bias in combined Pearson rs and Fisher’s z transformations. J. Gen. Psychol. 125 (3), 245–261. Cravalho, M.A., 1996. Toast on ice: the ethnopsychology of the winter-over experience in Antarctica. Ethos 24 (4), 628–656. Doll, R.E., Gunderson, E.K.E., 1971. Group size, occupational status and psychological symptomatology in an extreme environment. J. Clin. Psychol. 27 (2), 196–198. Dunlap, W.P., Jones, M.B., Bittner, A.C., 1983. Average correlations vs. correlated averages. Bull. Psychon. Soc. 21 (3), 213–216. Ellwardt, L., Aartsen, M., Deeg, D., Steverink, N., 2013. Does loneliness mediate the relation between social support and cognitive functioning in later life? Soc. Sci. Med. 98, 116–124. Francis, G., Bishop, L., Luke, C., Middleton, B., Williams, P., Arendt, J., 2008. Sleep during the Antarctic winter: preliminary observations on changing the spectral composition of artificial light. J. Sleep Res. 17, 354–360. Fredrickson, B.L., Losada, M.F., 2005. Positive affect and the complex dynamics of human flourishing. Am. Psychol. 60 (7), 678. Furnham, A., Stringfield, P., 1998. Congruence in job-performance ratings: a study of 360 feedback examining self, manager, peers, and consultant ratings. Hum. Relat. 51 (4), 517–530. Gavalas, A., 2011. Researching team dynamics, personality and behaviour of team members working on Antarctic research stations: a literature review and suggestions for a future study. OKS Rev. 1 (1), 1–12. Gierveld, J.D.J., Van Tilburg, T., 2006. A 6-item scale for overall, emotional, and social loneliness confirmatory tests on survey data. Res. Aging 28 (5), 582–598. Godwin, J.R., 1986. A Preliminary Investigation into Stress in Australian Antarctic Expeditioners: DTIC Document. Goldberg, L.R., 1981. Language and individual differences: the search for universals in personality lexicons. Rev. Personal. Soc. Psychol. 2 (1), 141–165. Grant, I., Eriksen, H.R., Marquis, P., Orre, I.J., Palinkas, L.A., Suedfeld, P., Ursin, H., 2007. Psychological selection of Antarctic personnel: the ’SOAP’ instrument. Aviat. Space Environ. Med. 78, 793–800. Gunderson, E.K.E., Nelson, P.D., 1963. Adaptation of small groups to extreme environments. Aero. Med. 34 (12), 1111–1115. Hawkley, L.C., Preacher, K.J., Cacioppo, J.T., 2010. Loneliness impairs daytime functioning but not sleep duration. Health Psychol. 29 (2), 124. Jenkins, D., Palmer, S., 2003. A review of stress, coping and positive adjustment to the challenges of working in Antarctica. Int. J. Health Promot. Educ. 41 (4), 117–131. John, O.P., Srivastava, S., 1999. The Big Five trait taxonomy: history, measurement, and theoretical perspectives. Handb. Personal.: Theory Res. 2, 102–138. Judge, T.A., Higgins, C.A., Thoresen, C.J., Barrick, M.R., 1999. The big five personality traits, general mental ability, and career success across the life span. Pers. Psychol. 52 (3), 621–652. Larsen, J.T., Norris, C.J., McGraw, A.P., Hawkley, L.C., Cacioppo, J.T., 2009. The evaluative space grid: a single-item measure of positivity and negativity. Cognit. Emot. 23 (3), 453–480. Long, C.R., Averill, J.R., 2003. Solitude: an exploration of benefits of being alone. J. Theory Soc. Behav. 33 (1), 21–44. Lopes, P.N., Brackett, M.A., Nezlek, J.B., Schütz, A., Sellin, I., Salovey, P., 2004. Emotional intelligence and social interaction. Personal. Soc. Psychol. Bull. 30 (8), 1018–1034. Lopes, P.N., Salovey, P., C^ ot�e, S., Beers, M., Petty, R.E., 2005. Emotion regulation abilities and the quality of social interaction. Emotion 5 (1), 113. Marshall, N.J., 1974. Dimensions of privacy preferences. Multivar. Behav. Res. 9 (3), 255–271. McAdams, D.P., 1980. A thematic coding system for the intimacy motive. J. Res. Personal. 14 (4), 413–432. McCormick, I.A., Taylor, A.J., Rivolier, J., Cazes, G., 1985. A psychometric study of stress and coping during the International Biomedical Expedition to the Antarctic (IBEA). J. Hum. Stress 11 (4), 150–156. McGorry, S.Y., 2000. Measurement in a cross-cultural environment: survey translation issues. Qual. Mark. Res. Int. J. 3 (2), 74–81. Nagar, D., Pandey, J., 1987. Affect and performance on cognitive task as a function of crowding and noise. J. Appl. Soc. Psychol. 17 (2), 147–157. Nelson, P.D., Gunderson, E.K.E., 1962. Analysis of Adjustment Dimensions in Small Confined Groups: DTIC Document. Nicolas, M., Bishop, S.L., Weiss, K., Gaudino, M., 2016. Social, occupational, and cultural adaptation during a 12-month wintering in Antarctica. Aero. Med. Hum. Perform. 87 (9), 781–789. Nutt, D., Wilson, S., Paterson, L., 2008. Sleep disorders as core symptoms of depression. Dialogues Clin. Neurosci. 3 (10), 329–336. Palinkas, L.A., 1992. Going to extremes: the cultural context of stress, illness and coping in Antarctica. Soc. Sci. Med. 35 (5), 651–664.
5. Conclusion Despite its limitation of sample size, the present study offers a theoretically-driven model tested using a culturally diverse group of participants. The data represent an agglomerate from five Antarctic stations from different NAP, contrasting with the usual studies con ducted in this field, usually focusing on one station or one programme. The results emphasise the importance of social adjustment to Antarctic stations and the relevance of using measures of social needs in the recruitment process. The study also stresses the central role isolation plays in one’s adjustment and its potential effect on cognitive func tioning, mood and job satisfaction. Declaration of competing interest None. Acknowledgement The authors would like to thank the Polar Research Institute of China, the Alfred Wegener Institute, Antarctica New Zealand, and the Norwegian Polar Institute for granting access to their winter-overs, as well as the participants who gave some of their free time while in Antarctica to make this study possible. References Agho, A.O., Price, J.L., Mueller, C.W., 1992. Discriminant validity of measures of job satisfaction, positive affectivity and negative affectivity. J. Occup. Organ. Psychol. 65 (3), 185–195. Arvey, R.D., Murphy, K.R., 1998. Performance evaluation in work settings. Annu. Rev. Psychol. 49 (1), 141–168. Beaton, D.E., Bombardier, C., Guillemin, F., Ferraz, M.B., 2000. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine 25 (24), 3186–3191. � Bishop, S., Lapierre, J., 2010. Human factors Binsted, K., Kobrick, R.L., Griofa, M.O., research as part of a Mars exploration analogue mission on Devon Island. Planet. Space Sci. 58 (7), 994–1006.
6
C. Jaksic et al.
Polar Science 22 (2019) 100484 Solignac, A., 2004. Influence des moyens de communication sur l’adaptation en situation d’isolement: Le cas d’un groupe d’hivernants en Terre Ad�elie [Influence of communcation channels on adjustment in an isolation context: Case of winter-overs in Ad�elie land]. Universit�e de Reims Champagne-Ardenne. Stewart, W.P., Cole, D.N., 2001. Number of encounters and experience quality in Grand Canyon backcountry: consistently negative and weak relationships. J. Leis. Res. 33 (1), 106. Strange, R., Klein, W., 1973. Emotional and social adjustment of recent US winter-over parties in isolated Antarctic stations. In: Edholm, O.G., Gunderson, E.K.E. (Eds.), Polar Human Biology: Proceedings of the SCAR/IUPS/IUBS Symposium on Human Biology and Medicine in the Antarctic. Heinemann, Chicago, pp. 410–416. Taylor, A.J.W., McCormick, I.A., 1987. Reactions of family partners of Antarctic expeditioners. Polar Rec. 23 (147), 691–700. Thompson, E.R., Phua, F.T., 2012. A brief index of affective job satisfaction. Group Organ. Manag. 37 (3), 275–307. van Winkel, M., Wichers, M., Collip, D., Jacobs, N., Derom, C., Thiery, E., Peeters, F., 2017. Unraveling the role of loneliness in depression: the relationship between daily life experience and behavior. Psychiatry 2 (80), 104–117. Vaughan, G.M., Hogg, M.A., 2014. Social Psychology: French Forest. Pearson Australia, N. S.W. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Personal. Soc. Psychol. 54 (6), 1063. Westin, A.F., 1967. Privacy and Freedom. Atheneum, New York. Westin, A.F., 2003. Social and political dimensions of privacy. J. Soc. Issues 59 (2), 431–453. Winter, D.G., John, O.P., Stewart, A.J., Klohnen, E.C., Duncan, L.E., 1998. Traits and motives: toward an integration of two traditions in personality research. Psychol. Rev. 105 (2), 230. Zeedyk-Ryan, J., Smith, G.F., 1983. The effects of crowding on hostility, anxiety, and desire for social interaction. J. Soc. Psychol. 120 (2), 245–252.
Palinkas, L.A., 2002. On the ice: individual and group adaptation in Antarctica. Retrieved from. http://www.bec.ucla.edu/papers/Palinkas_On_The_Ice.pdf. Palinkas, L.A., Gunderson, E.K.E., Burr, R.G., 1989. Psychophysiological Correlates of Human Adaptation in Antarctica. U.S.A.: Naval Health Research Centre, San Diego, CA. Retrieved from. http://www.dtic.mil/cgi-bin/GetTRDoc?Location¼U2&do c¼GetTRDoc.pdf&AD¼ADA216679. Palinkas, L.A., Gunderson, E.K.E., Holland, A.W., Miller, C., Johnson, J.C., 2000. Predictors of behavior and performance in extreme environments: the Antarctic space analogue program. Aviat. Space Environ. Med. 71 (6), 619–625. Peri, A., Barbarito, M., Barattoni, M., Abraham, A., 2000. The dynamics and the interpersonal and intrapersonal relations within an isolated group in extreme environments. Small Group Res. 31, 251–274. Rona, R.J., Li, L., Gulliford, M.C., Chinn, S., 1998. Disturbed sleep: effects of sociocultural factors and illness. Arch. Dis. Child. 78 (1), 20–25. Rothmann, S., Coetzer, E.P., 2003. The big five personality dimensions and job performance. SA J. Ind. Psychol. 29 (1), 68–74. Roy, D.D., Deb, N., 1999. A Note on Job Satisfaction Level of Scientists and Logistic Personnel in Antarctic Expedition. National Centre for Antarctic and Ocean Research, Department of Ocean Development, New Delhi, India. Ryan, R.M., Deci, E.L., 2000. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am. Psychol. 55 (1), 68–78. Sandal, G.M., van deVijver, F., Smith, N., 2018. Psychological hibernation in Antarctica. Front. Psychol. 9, 2235. Sch€ onbrodt, F.D., Gerstenberg, F.X., 2012. An IRT analysis of motive questionnaires: the unified motive scales. J. Res. Personal. 46 (6), 725–742. Silver, N.C., Dunlap, W.P., 1987. Averaging correlation coefficients: should Fisher’s z transformation be used? J. Appl. Psychol. 72 (1), 146. Smith, S., Haythorn, W.W., 1972. Effects of compatibility, crowding, group size, and leadership seniority on stress, anxiety, hostility, and annoyance in isolated groups. J. Personal. Soc. Psychol. 22 (1), 67.
7
Contents lists available at ScienceDirect
Polar Science journal homepage: http://www.elsevier.com/locate/polar
Antarctic stations as workplaces: Adjustment of winter-over crew members Cyril Jaksic, Gary Steel *, Emma Stewart, Kevin Moore Department of Tourism, Sport & Society Faculty of Environment, Society & Design P O Box 85084 Lincoln University, Lincoln, 7647, Christchurch, New Zealand
A R T I C L E I N F O
A B S T R A C T
Keywords: Antarctica Isolated and confined environments Extreme environment Psychological Adaptation Person-environment fit Polar psychology
The living conditions at Antarctic stations can be challenging for support personnel. It has been; suggested that the experience of isolation and confinement can contribute to the emergence of the; winter-over syndrome. The present study adopts a Person-Environment fit approach to investigate; individual adjustment to the social constraints of an Isolated and Confined Environment (ICE). The; study gathered monthly data from 14 partici pants from five different stations, run by different National Antarctic Programmes. Results revealed that a lack of privacy generated by the confinement is associated with sleep disturbance. In addition, a high level of loneliness, experienced as a result of the; isolation, is positively related to cognitive impairment and negatively related to job satisfaction and; positive/negative mood ratio. The results further suggest that loneliness can be predicted by a predeployment; measure of need for affiliation, as well as levels of the personality traits of agreeableness; and extraversion.
1. Introduction Human activity in Antarctica has typically been predominantly focused on scientific research. Most Antarctic stations, which offer the facilities needed for such research, are generally operated by National Antarctic Programmes (NAP). Each NAP is responsible for selecting the science and support personnel that will be deployed to their station. The ability to adjust in a positive way to such environments varies from individual to individual (Peri et al., 2000; Roy and Deb, 1999). It is, therefore, important to recruit people who will best fit with critical features of this environment in order to increase the likelihood of pos itive outcomes, both professional and personal. The polar environment presents its own set of unique features in regards to human psychology. Those going to Antarctic stations have to deal with the remoteness of the continent. This remoteness places con straints on the number and types of interactions that can occur with the world north of 60� S with sometimes limited phone or internet access. Such social isolation has been noted as being amongst the major stressors in Antarctica (McCormick et al., 1985). Separation from family and friends was also the most frequently mentioned source of stress for two expeditions that spent 17 days in complete isolation in the Arctic (Bishop et al., 2001). It was also ranked as the main stressor by different Australian crews wintering-over in Antarctica between 1980 and 1982 (Godwin, 1986). Strange and Klein (1973) found that almost all anxious feelings reported by U.S. personnel over the winter were related to
events occurring back home. It is not helping that perceived peer sup port degrades over the winter-over (Nicolas et al., 2016). It is important to note that while the term ‘isolation’ might refer to a physical description (being physically isolated), it only becomes a problem when the isolation from the usual social network is negatively experienced and leads one to feel lonely. It is acknowledged that isola tion does not systematically lead to feeling lonely, nor that it is a necessary precondition for it, but rather it can facilitate the emergence of such a feeling. The distinction between isolation and loneliness is important because it is possible for two individuals to be in the same physically isolated environment but not feel equal levels of loneliness; for some, it may even be a motivating factor for working in such a re gion. The present paper will not consider the objective isolation but, instead, the negative experience of a perceived isolation, in other words; loneliness. In addition to loneliness, confinement can also take its toll on Ant arctic personnel. They must cope with having to share living spaces with workmates, scientists, and VIP visitors, usually in very close quarters. This includes sharing facilities such as bathrooms, dining halls, and leisure rooms. During the busiest season (summer), personnel usually share bedrooms. There are often limited opportunities to venture beyond the immediate confines of the station, especially in winter due to extreme weather or other safety considerations. The effects that these psychological challenges may generate are even more prevalent for those who spend an entire winter in an Antarctic station; so-called
* Corresponding author. E-mail address: [email protected] (G. Steel). https://doi.org/10.1016/j.polar.2019.100484 Received 29 April 2019; Received in revised form 26 September 2019; Accepted 16 October 2019 Available online 18 October 2019 1873-9652/© 2019 Elsevier B.V. and NIPR. All rights reserved.
C. Jaksic et al.
Polar Science 22 (2019) 100484
winter-over personnel. This confinement may lead to experiencing a lack of privacy. In one study, perceived lack of privacy has been reported as the second most important source of stress by Australian winter-overs (Godwin, 1986). Cravalho (1996) mentions the case of a woman who complained that even when she wanted to get away from McMurdo station, it had to be done with someone else. As noted by Binsted et al. (2010), this concern over privacy also applies to the limited auditory privacy due to the layout of polar stations. Finally, as one winter-over reported, in an Antarctica station “tu entends tout. En fait tu n’as aucune intimit� e” [you hear everything. Actually, you have no privacy] (Solignac, 2004, p. 130). It has been suggested (Palinkas, 1992, 2002) that both loneliness and confinement could be partially responsible for commonly observed symptoms amongst winter-overs, clustered together under the name of the winter-over syndrome. The winter-over syndrome is characterised by negative mood (depression and hostility), sleep disturbance and impaired cognition (Strange and Klein, 1973). The notion that isolation and confinement, and its potentially concomitant loneliness and lack of privacy, can contribute to the emergence of the winter-over syndrome is supported by studies conducted outside of Antarctica. For instance, it has been shown that social isolation was related to sleep disturbance (Cacioppo et al., 2000; Hawkley et al., 2010) and cognitive impairment (e.g., memory, attention, information processing speed) (Ellwardt et al., 2013). Further, indirect evidence for the link between confinement and the same outcome variables is provided by research on experiences of crowding. It has also been found that crowding is related to negative mood (Nagar and Pandey, 1987; Smith and Haythorn, 1972; Zee dyk-Ryan and Smith, 1983), sleep disturbance (Rona et al., 1998) and cognitive impairment (Nagar and Pandey, 1987). Both crowding and issues of privacy arise when human density is perceived as higher than desired. In addition, feeling crowded conceptually entails that one does not have adequate privacy. Given these findings, it seems plausible that both isolation and confinement might be contributing to the emergence of the winter-over syndrome. Isolation and confinement can also adversely affect a whole team, and even the NAP itself. For instance, at least one field accident has been attributed to the distraction caused by the first mail delivery from an Antarctica expeditioner’s family (Taylor and McCormick, 1987). Pal inkas (2002) reported the case of two men who had to be sent back home before the end of their mission when preoccupations with their family prevented them from efficiently carrying out their assignment. For all NAP, sending individuals to the Ice and keeping the stations running is expensive. Replacing an employee can be costly and logistically difficult. Therefore, adjusting to this unusual environment is crucial for the in dividual’s well-being, the whole team, and the NAP. In order to cope with the lack of privacy, Jenkins and Palmer (2003) have proposed that social withdrawal could be an effective regulation strategy. It is assumed that, to a certain extent, the negative conse quences of a lack of privacy can be thwarted by some strategies, either cognitive (e.g., mentally shutting out other individuals) or behavioural (e.g., physically withdrawing in one’s room or outside). As an example, Cravalho (1996) reported that a crew member started working nights to limit social interactions. The lack of privacy must, therefore, be considered in conjunction with the strategies put in place to preserve one’s privacy. To investigate adjustment to an Antarctic station in this study, a specific theoretical approach was adopted: Person-Environment fit theory (henceforth referred to as P-E fit). This theory is based on the assumed interdependence of an individual’s and an environment’s characteristics. It posits that, for relevant environmental features, the match with one individual’s corresponding characteristics can predict general adjustment to that environment. In Antarctica, therefore, the extent to which one adjusts to the isolation and confinement should predict the extent of one’s winter-over syndrome, job performance and job satisfaction.
Because the major issue with isolation is the potential frustration of one’s need for directly socialising with one’s normal social network, it is expected that needs associated with social relationships are the most relevant predictors of one’s adjustment to isolation. A number of such needs have been identified (see, e.g., Ryan and Deci, 2000). We consider here the need for affiliation and the need for intimacy. The need for affiliation has been defined as “establishing relationships €nbrodt and Gersten to rather unfamiliar people and acquaintances” (Scho berg, 2012, p. 4) and as “the urge to form connections and make contact with other people” (Vaughan and Hogg, 2014, p. 503). The need for in timacy is concerned with “being close to others, having positive profound interactions, and practicing self-disclosure and warm mutual exchange” €nbrodt and Gerstenberg, 2012, p. 4) and “experiencing a warm, (Scho close, and communicative exchange with another person” (McAdams, 1980, p. 413). The critical differences between the two concepts are those of depth and closeness. The need for intimacy pushes one to seek re lationships that manifest these qualities, whereas the need for affiliation simply motivates one to make social contacts, however superficial. Taken together, though, it is expected that higher levels of need for affiliation and need for intimacy will put a greater demand on what can be provided in the social environment, thus increasing the likelihood of loneliness in an Antarctic setting. The degree of confinement occasionally experienced in an Antarctic station can interact with the need for privacy and the need for intimacy. Confinement, that potentially leads to a lack of privacy, will present greater challenges to an individual with a high need for privacy. Simi larly, a high need for intimacy requires that individuals have sufficient privacy to bond and develop a close relationship with one or more other people. As a result, confinement will be more challenging for individuals with high need for intimacy. However, the relationship between one’s needs and the extent to which they manifest themselves is not straightforward. The channelling hypothesis (Winter et al., 1998) suggests that, while the needs determine our goals, personality traits can affect the means used to reach them. For instance, two individuals with a high need for intimacy but differing levels of extraversion will not necessarily behave the same way in the same context. One who is extraverted might feel comfortable making many relationships in the hope that a few will provide intimacy. In contrast, an introvert might choose to spend time with only a few people but will put extra effort into creating intimacy. With the focus of this study on social needs predicting social adjustment, two personality traits from the Big Five (Goldberg, 1981) have been identified as potentially influential channels, namely agreeableness and extraversion. Those two personality traits have been systematically found to correlate with social variables (e.g., social support, interpersonal competence) (Judge et al., 1999; Lopes et al., 2004; Lopes et al., 2005). Moreover, they are the only two traits that are fundamentally defined by the way we interact with others (extraversion; sociability, assertiveness, talkativeness/agree ableness; altruistic, sympathetic, helpful) (Rothmann and Coetzer, 2003). On the other hand, neuroticism relates to how one deals with negative affects, conscientiousness relates to approaching and carrying out tasks and openness to experience relates to one’s imagination and curiosity. None of which specifically deals with social environments. Therefore, agreeableness and extraversion are the most relevant traits for predicting people’s responses to the unique social environment found in Antarctic stations. The present study aimed at testing a theoretical model taking a P-E fit approach and tailored to two of the most relevant Antarctic social fea tures. The first part of the model predicts a positive relationship between pre-deployment measures of need for affiliation and intimacy with loneliness. The second part predicts a negative relationship between measures of need for intimacy and privacy with adjustment with the lack of privacy. The relationships in both parts are expected to be moderated by both extraversion and agreeableness. In addition, both loneliness and the fit with privacy are expected to be related to job satisfaction, job performance, cognitive impairment, sleep disturbance and mood. 2
C. Jaksic et al.
Polar Science 22 (2019) 100484
2. Method
(e.g., talkative, reserved, inventive). The participants had to assess whether each attribute describes them using a 5-point Likert-type response format. To record the importance of the need for social contact, each partic ipant completed two subscales from the Unified Motives Scales €nbrodt and Gerstenberg, 2012): the need for affiliation and need (Scho for intimacy scales. Respondents were asked to evaluate 10 affiliation and 10 intimacy items using a 5-point Likert-type response format. In order to measure the need for privacy, the Preference for Solitude Scale (Burger, 1995) was used. This questionnaire proposes 12 pairs of statements (e.g., Time spent alone is often productive for me vs Time spent alone is often time wasted for me) from which respondents chose the one that described them the best. Though the Preference for Solitude Scale measures the inclination toward solitude, Long and Averill (2003) and Westin (2003) report that solitude is a type of privacy. Westin (1967, p. 31) even considers solitude as “the most complete state of privacy that individuals can achieve”. Other researchers have similarly regarded the two concepts as being part of the same construct (Stewart and Cole, 2001). Hence, it is used in this study as a proxy measure for need for privacy. A high preference for solitude should reflect a high need for privacy.
2.1. Recruitment Fifteen NAP that have at least one permanent station in Antarctica were contacted via email in 2015. These included: Argentina, Australia, Chile, China, France, Germany, India, Japan, New Zealand, Norway, Russia, South Africa, South Korea, the United Kingdom and the United States. Those which agreed to forward the advertisement for the present study to their future winter-overs were: China, Germany, New Zealand and Norway. However, because the New Zealand Programme only gave approval in early 2016, once their future winter-overs were already at the station, the data for their first months at the station are missing. The other NAP either failed to reply despite receiving reminders or simply explained they could not be involved in the study for diverse reasons. Despite the absence of response from the U.S. NAP, one of their future winter-overs personally contacted the researcher to take part in the study. This person had read an interview the first author had given about the study and that had been published online. 2.2. Sample Amongst the 57 winter-overs employed by the four collaborating NAP, a total of 14 (26%) of the winter-overs filled in the pre-winter-over survey and at least one monthly survey. Amongst those 14 participants, there was one in Great Wall Station (China), four at Neumayer Station III (Germany), three at Scott Base (New Zealand), five at Troll Station (Norway) and one at Amundsen-Scott South Pole Station (U.S.). There was a total of six females and eight males. With one male failing to report his age, the average age of the 13 remaining participants was 35.7 years old (SD ¼ 8.9), ranging from 24 to 52 years old. Out of the 14 partici pants, three reported having previously wintered-over at least once, 10 had never wintered-over previously, and one did not respond to this item. All participants took part in the present study during a 2016 winter-over. This research was approved by the Lincoln University Human Ethic Committee.
2.5. Monthly survey measures The second survey was completed online on a monthly basis by the winter-overs throughout their stay in Antarctica. It comprised questions directly related to the participants’ experiences at their respective sta tions for the previous 30 days. A repeated measures was adopted in order to test the existence of correlations between the variables of in terest throughout the year. The intention was not to investigate the ef fect of time but instead, to test that relationships exist overall. Making measures at only one point in time might reveal relationships that only exist at such a specific moment. Revealing significant relationships be tween variables measured throughout the year would strengthen the results and make it more generalizable to Antarctic stays in their entirety. Job satisfaction was measured through the Brief Index of Affective Job Satisfaction (BIAJS) (Thompson and Phua, 2012). This question naire asks one to assess, on a 5-point Likert-type scale, how strongly one agrees with four statements (e.g., Most days I am enthusiastic about my job). Sleep quality was assessed via three items that have already been used in an Antarctic context (Doll and Gunderson, 1971; Palinkas et al., 1989). For the three of them (difficulties falling asleep or staying asleep; waking up at night; and feeling tired during the day), the participant had to indicate the frequency they experienced them in the past thirty days answering on a 5-point Likert-type response format. Positive and negative moods are not simply two ends of a continuum but, rather, two independent constructs (Agho et al., 1992; Watson et al., 1988). For this reason, the same format of response as the Evaluative Space Grid (ESG) by Larsen et al. (2009) was used. This single-item measure consists of a 5 � 5 grid with the x-axis representing positive affect and the y-axis representing negative affect. Each scale proposes five options (not at all; slightly; moderately; quite a bit; extremely). The participants were asked to indicate, by choosing one square, the extent to which they felt positive and negative affect during the last 30 days. This measure allows calculation of a ratio between negative and positive affect. Because it has been suggested that negative affect has a weighting of about three times more than positive affect in its effect on one’s overall well-being (Fredrickson and Losada, 2005), the ratio calculated divides the positive affect score by three times the negative affect score and is labelled Mood Ratio. Cognitive impairment was assessed via four items covering four different cognitive functions, namely memory, attention, language, and thinking. The items were as follows: Over the last 30 days, how often have you experienced: 1) being forgetful 2) difficulties to focus 3)
2.3. Procedure While most studies take pre- and post-winter-over measures (Gav alas, 2011), such an approach would not be sensitive enough to capture the variations that occur in one’s adjustment during the course of a person’s stay. It was felt that a self-report score obtained at the end of the stay was likely to represent an overall appreciation of the experience. Furthermore, such answers could be unduly influenced by the cognitive and affective state at the time of reporting, which might not validly reflect the entire experience throughout the winter. For this reason, a more fine-grained approach has been used here. The study consisted of two different surveys. The first was a pre-winter-over survey that par ticipants had to complete at any time prior to their arrival in Antarctica. However, because the New Zealand programme agreed to collaborate one month after their winter-overs arrived in Antarctica, those three participants filled in the pre-winter-over survey while on the ice. For each participant, a code was generated and provided to them every time a survey link would be sent. This code was for them to enter in each survey. This allowed for participants to fill in questionnaires anony mously while allowing the researchers identify the data collected over time as coming from the same individual. 2.4. Pre-winter-over measures The pre-winter-over survey consisted of demographic questions and the following measures. Personality traits were measured using the Big Five Inventory (BFI) (John and Srivastava, 1999). This questionnaire presents 44 attributes 3
C. Jaksic et al.
Polar Science 22 (2019) 100484
difficulties to find your words or express what you meant 4) feeling confused. The participant answered using a 5-point Likert-type response format. One overall score was calculated by averaging the four answers. P-E fit in regards to isolation was measured by asking participants to complete an adapted version of the 6-item De Jong Gierveld Loneliness Scale (Gierveld and Van Tilburg, 2006). It presents six statements to which participants respond using a 5-point Likert-type response format. In effect, this produces a measure of ‘non-fit’ with the social environ ment, where one’s needs and desires regarding social interactions are left unsatisfied. This means that a low score on Loneliness reflects a good P-E fit. Because of this, this variable will be referred to as Loneliness from this point onwards. To measure P-E fit for privacy, the concept of privacy was first defined. Participants were then asked to indicate the extent to which they had enough privacy, using a 5-point Likert-type response format. In addition, they were asked what behaviour they used to regulate their privacy. They were also asked to report the frequency with which each privacy behaviour was used. However, because the measure of privacy regulation yields non-numerical data, it will not be further addressed in the present paper but will be presented in a separate publication, and scores of P-E fit for privacy will be taken from the single item measure. To assess one’s job performance in Antarctica, peer-evaluation has often been used. Such measures can show high degrees of agreement amongst peers (Nelson and Gunderson, 1962), and it is considered a valid measure of job performance in ICEs (Grant et al., 2007; Gunderson and Nelson, 1963; Palinkas et al., 2000) and in more mundane work places (Arvey and Murphy, 1998; Furnham and Stringfield, 1998). The participants were asked to name, based on their job performance, the individuals at their station they would choose if they had to recruit a team for a future expedition. For each nominee, participants had to indicate their level of acquaintance. In order to accommodate non-English speaking participants, both surveys were translated into Chinese and German. Both versions had been back-translated to ensure the accuracy of the translation. Backtranslation has been considered as a good approach to ensure accu racy in translation (McGorry, 2000), and though alternative methods have been deemed more accurate (Beaton et al., 2000), they present the disadvantage of being extremely time-consuming and costly.
In addition, other participants simply did not respond to this item. Therefore, the numerous missing data on that measure were removed from the analyses, which left no measure of Job Performance. All existing scales used in this study had been checked for internal validity and reliability by their respective authors and were found to meet the relevant criteria. Scales created by the authors of the present study (Sleep Disturbance and Cognitive Impairment) were found to have an internal consistency (Cronbach’s α) above .80. The descriptive sta tistics for the eleven variables can be found in Table 2 below. For repeatedly measured variables, an average score was first calculated for each participant and then, the mean and standard deviation across the 14 participants was computed. 3.2. Analysis Two distinct methods could have been used to analyse the data with regard to the hypotheses. First, averaging correlations consists of calculating, for each month, the correlations of two variables of interest. This method results in 12 correlations (one per month) that are averaged to have one value reflecting the overall relationship between the two variables across time. The second method consists of correlating aver ages. Here, for each month, the average values of two variables of in terest are calculated. This method results in 12 averages per variable (one per month) that can then be correlated across the 12 months. It has been demonstrated that averaging correlations is more reliable and less biased than correlating averages (Bittner et al., 1982; Dunlap et al., 1983). For this reason, the data in this study were analysed using the averaging correlations method. However, while averaging correlations appears to be the best option, it has been found that averaging raw correlations produces greater bias than averaging Fisher’ z-transformed correlations (Silver and Dunlap, 1987). The advantage of the latter method appears to be even greater when the sample size is small (Corey et al., 1998). For each month, the Spearman correlation between two variables of interest was calculated. The non-parametric Spearman correlations were chosen over the parametric Pearson correlations because the small sample size made it difficult to test for the normality of the data, which is Table 2 Descriptive statistics for all eleven variables.
3. Results 3.1. Data
Variables
Means
Standard deviations
Minimum
Maximum
Scale score range
Table 1 shows the number of participants who completed a survey in each of the 13 months of the study. Since the concern was not with objective time (e.g., how people feel in May) but rather with relative time (e.g., how people feel during their second month at the station), and because not every participant arrived in Antarctica at the same time, each winter-over’s timeframe was aligned to start with their respective arrival month. Some participants failed to fill in their monthly survey, therefore almost all months had missing data. Only two participants filled in a survey after 13 months of stay. Therefore, only the first 12 months of each participant’s responses were taken into consideration for analyses. Out of the 14 participants, one Chinese and one US-American participants were the only winter-overs at their respective station to take part in the study. As a result, the peernomination method to assess job performance did not occur for them.
Need for Affiliation Need for Intimacy Need for Privacy Agreeableness Extraversion Job Satisfaction Sleep Disturbance Mood Ratio Cognitive Impairment Loneliness Privacy Fit
2.91
0.36
2.20
3.40
1–5
2.76
0.69
1.50
3.90
1–5
6.71
1.82
3.00
9.00
0–12
3.85 3.20 3.99
0.52 0.48 0.56
3.11 2.38 2.89
4.89 3.88 4.97
1–5 1–5 1–5
2.54
0.86
1.12
4.14
1–5
0.73 2.15
0.35 0.72
0.41 1.04
1.67 3.50
0.07–1.67 1–5
2.20 4.17
0.50 0.57
1.29 2.82
3.35 4.86
1–5 1–5
N ¼ 14.
Table 1 Number of participants (n) who filled in their survey after each month spent in Antarctica. months n
1
2
3
4
5
6
7
8
9
10
11
12
13
7
13
10
12
12
12
12
14
12
11
10
9
2
4
C. Jaksic et al.
Polar Science 22 (2019) 100484
assumed by Pearson correlations. For each pair of variables of interest, 12 Spearman correlations (one per month) were obtained. Once the 12 Spearman correlations between two variables of interest were computed, they were converted using the Fisher’s z transformation. The average and standard deviation of those z scores were then obtained. Both the average and the standard deviation have been backtransformed to be expressed as correlation coefficients, thus making it easier to assess the effect size. This average was then tested against zero using a t-test to find out if the averaged correlation is significant. As testing of the model produced 18 individual correlations, a Bon ferroni correction was required to reduce the chance of a Type I error (false positive). As a result, the level of significance is divided by 18, and all the p-values presented below should be compared to an α ¼ .003125 for determining significance at the p < .05 level.
Table 4 Back-transformed Spearman correlations after a Fisher’s z transformation be tween Loneliness and Privacy Fit, and the relevant outcome variables (all df ¼ 11). Correlations with
Loneliness
Privacy Fit
3.2.1. Predicting P-E fit Due to the small sample size, it was not possible to test for the moderation effect of Agreeableness and Extraversion on the relationship between the three needs and the two measures of P-E fit. Hence, the two personality traits were taken as simple predictors of Loneliness and Privacy Fit. Both personality traits were expected to have a negative relationship with both Loneliness and Privacy Fit. First, the extent to which both P-E fits (Loneliness and Privacy Fit) were predicted by the pre-winter-over measures was examined. Table 3 shows the correlations between the two measures of P-E fit and their respective predictors. As expected, Loneliness positively correlated with the Need for Affiliation and negatively correlated with Agreeableness. However, Loneliness and the Need for Intimacy were not significantly correlated. Moreover, Loneliness was positively correlated with Extraversion, where a negative relationship was expected. As predicted, the Privacy Fit did negatively correlate with the Need for Intimacy. However, there was no significant correlation between Privacy Fit and Need for Privacy, Agreeableness, and Extraversion. All significant correlations showed modest associations between the variables.
rS r2S p-value (1tailed) rS r2S p-value (1tailed)
Job Satisfaction
Sleep Disturbance
Cognitive Impairment
Mood Ratio
-.487 .237 <.001
.219 .048 n.s.
.486 .236 <.001
-.468 .219 <.001
.190 .036 n.s.
.404 .163 <.001
.208 .043 n.s.
-.04 .002 n.s.
lack of privacy as being an issue (Godwin, 1986), it does not directly relate to aspects of adjustment. So, while concerns are expressed about the lack of privacy in ICEs (Binsted et al., 2010), it is possible that its consequences have less impact on adjustment than was previously hypothesised. Although this seems – at first glance – somewhat unlikely, it may simply be that winter-overs become aware of a lack of privacy only when they are asked about it. This could be interpreted as a sign of adaptation, inasmuch as the condition has become “usual” and, there fore, unremarkable, and of less influence on day-to-day living. This is supported by a recent study finding the use of withdrawing strategies to decrease over time (Sandal, van deVijver and Smith, 2018). The authors suggest that winter-overs might become emotionally dumb across time and thus, their situational awareness decreases. Alternatively, it is conceivable that privacy is of considerable importance but that the measure used in the present study failed to capture the construct accurately. While the item measuring Privacy Fit grossly defined privacy as one general concept, it might be beneficial to analyse the measure into its constituent parts. Privacy has been identi fied as having different components such as anonymity (not being identified in public) or not neighbouring (relationship with neighbours) (Marshall, 1974). It might therefore be important for a measure of pri vacy to cover the full range of dimensions that the construct encompasses. There are well-established links between loneliness and depression (Nutt et al., 2008) and between depression and sleep disturbances (van Winkel et al., 2017). Thus, we expected to see a relationship between Loneliness and Sleep Disturbance but we did not find any reliable evi dence of this. The link may not be as strong as was supposed, or its absence in our study may be due to environmental features. We know that sleep impairment can be partially caused by the disruption of the circadian cycle via the unusual light-cycle in Antarctica (Francis et al., 2008). It is possible that the effect of loneliness on sleep impairment is overshadowed by this factor, which masks the more subtle relationship. A larger number of participants and controlling for the effects of the light-cycle may bring the relationship to the fore. Placing loneliness at the core of model, rather than as simply an outcome variable, allowed it to be identified as a central factor relevant to cognitive functioning, job satisfaction and mood. This suggests that
3.2.2. P-E fits predicting outcome variables The second step of the analysis consisted of looking at the relation ship between the two P-E fit measures and outcome variables relevant for winter-overs, namely Job Satisfaction, Sleep Disturbance, Cognitive Impairment and Mood Ratio. Table 4 shows the correlations between Loneliness and the outcome measures. As predicted, Loneliness positively correlated with Cognitive Impairment and negatively with Job Satisfaction and Mood Ratio. However, no significant correlation was found between Loneliness and Sleep Disturbance. Contrary to expectation, a significant positive cor relation has been found between Privacy Fit and Sleep Disturbance. No significant correlation was found between Privacy Fit and Job Satis faction, Cognitive Impairment, and Mood Ratio. As in the previous table, all significant correlations showed modest associations between the variables. 4. Discussion The fact that Privacy Fit did not correlate as expected with any of the outcome variables could suggest that, though winter-overs report the
Table 3 Back-transformed Spearman correlations after a Fisher’s z transformation between Loneliness and Privacy Fit, and their respective predictors (all df ¼ 11). Loneliness Privacy Fit
rS r2S p-value (1-tailed) rS r2S p-value (1-tailed)
Need for Affiliation
Need for Intimacy
.252 .064 <.001
.052 .003 n.s. -.319 .102 .002
5
Need for Privacy
Agreeableness
Extraversion
-.129 .017 n.s.
-.228 .052 .002 008 <.001 n.s.
.278 .077 <.001 -.223 .050 n.s.
Polar Science 22 (2019) 100484
C. Jaksic et al.
NAP should pay particular attention to future winter-overs’ needs for social contact. The present study suggests that those who adjust well with the isolation, and will not feel lonely as a result, will present better indices of affective, cognitive, and occupational adjustment. One way to combat the sense of isolation, and possibly loneliness, at an Antarctic station has been to increase the quality and quantity of communication with home. However, a recent study has suggested that this may make things worse for winter-overs. It was pointed out that some winter-overs spend so much time on the Internet to communicate with their relatives that they isolate themselves from the rest of the crew (Solignac, 2004). This can threaten group cohesion and efficiency, and make the experience of isolation quite unpleasant. It was also seen that some winter-overs focussed so much on the news received regularly from their friends and family that they neglected their work and needed to be sent back home (Palinkas, 2002) or even had field accidents because of the distraction it represented (Taylor and McCormick, 1987). Thus, a balance must be struck between contact with home and contact within the isolated group, taking into account the individual’s needs for affiliation. Based on the present study, we would recommend that NAP pay particular attention to the candidates’ needs for social contact during the recruitment process. On a final note, the small sample size of the present study is not uncommon when investigating winter-overs. This field of research has, therefore, relied on replications for strengthening the confidence in our knowledge. When similar relationships are systematically found, it suggests that the real relationships must be strong. This is especially true if said relationships are observed even with small sample sizes. More over, results found with culturally diverse samples and at different sta tions suggest that observed effects are widespread.
Bishop, S.L., Grobler, L.C., Schjøll, O., 2001. Relationship of psychological and physiological parameters during an Arctic ski expedition. Acta Astronaut. 49 (3), 261–270. Bittner, J.A.C., Dunlap, W.P., Jones, M.B., 1982. Averaged Cross-Correlations with Differentially-Stable Variables: Fewer Subjects Required with Repeated Measures. Sage Publications Sage CA, Los Angeles, CA. Symposium conducted at the meeting of the Proceedings of the Human Factors Society Annual Meeting. Burger, J.M., 1995. Individual differences in preference for solitude. J. Res. Personal. 29 (1), 85–108. Cacioppo, J.T., Ernst, J.M., Burleson, M.H., McClintock, M.K., Malarkey, W.B., Hawkley, L.C., et al., 2000. Lonely traits and concomitant physiological processes: the MacArthur social neuroscience studies. Int. J. Psychophysiol. 35 (2), 143–154. Corey, D.M., Dunlap, W.P., Burke, M.J., 1998. Averaging correlations: expected values and bias in combined Pearson rs and Fisher’s z transformations. J. Gen. Psychol. 125 (3), 245–261. Cravalho, M.A., 1996. Toast on ice: the ethnopsychology of the winter-over experience in Antarctica. Ethos 24 (4), 628–656. Doll, R.E., Gunderson, E.K.E., 1971. Group size, occupational status and psychological symptomatology in an extreme environment. J. Clin. Psychol. 27 (2), 196–198. Dunlap, W.P., Jones, M.B., Bittner, A.C., 1983. Average correlations vs. correlated averages. Bull. Psychon. Soc. 21 (3), 213–216. Ellwardt, L., Aartsen, M., Deeg, D., Steverink, N., 2013. Does loneliness mediate the relation between social support and cognitive functioning in later life? Soc. Sci. Med. 98, 116–124. Francis, G., Bishop, L., Luke, C., Middleton, B., Williams, P., Arendt, J., 2008. Sleep during the Antarctic winter: preliminary observations on changing the spectral composition of artificial light. J. Sleep Res. 17, 354–360. Fredrickson, B.L., Losada, M.F., 2005. Positive affect and the complex dynamics of human flourishing. Am. Psychol. 60 (7), 678. Furnham, A., Stringfield, P., 1998. Congruence in job-performance ratings: a study of 360 feedback examining self, manager, peers, and consultant ratings. Hum. Relat. 51 (4), 517–530. Gavalas, A., 2011. Researching team dynamics, personality and behaviour of team members working on Antarctic research stations: a literature review and suggestions for a future study. OKS Rev. 1 (1), 1–12. Gierveld, J.D.J., Van Tilburg, T., 2006. A 6-item scale for overall, emotional, and social loneliness confirmatory tests on survey data. Res. Aging 28 (5), 582–598. Godwin, J.R., 1986. A Preliminary Investigation into Stress in Australian Antarctic Expeditioners: DTIC Document. Goldberg, L.R., 1981. Language and individual differences: the search for universals in personality lexicons. Rev. Personal. Soc. Psychol. 2 (1), 141–165. Grant, I., Eriksen, H.R., Marquis, P., Orre, I.J., Palinkas, L.A., Suedfeld, P., Ursin, H., 2007. Psychological selection of Antarctic personnel: the ’SOAP’ instrument. Aviat. Space Environ. Med. 78, 793–800. Gunderson, E.K.E., Nelson, P.D., 1963. Adaptation of small groups to extreme environments. Aero. Med. 34 (12), 1111–1115. Hawkley, L.C., Preacher, K.J., Cacioppo, J.T., 2010. Loneliness impairs daytime functioning but not sleep duration. Health Psychol. 29 (2), 124. Jenkins, D., Palmer, S., 2003. A review of stress, coping and positive adjustment to the challenges of working in Antarctica. Int. J. Health Promot. Educ. 41 (4), 117–131. John, O.P., Srivastava, S., 1999. The Big Five trait taxonomy: history, measurement, and theoretical perspectives. Handb. Personal.: Theory Res. 2, 102–138. Judge, T.A., Higgins, C.A., Thoresen, C.J., Barrick, M.R., 1999. The big five personality traits, general mental ability, and career success across the life span. Pers. Psychol. 52 (3), 621–652. Larsen, J.T., Norris, C.J., McGraw, A.P., Hawkley, L.C., Cacioppo, J.T., 2009. The evaluative space grid: a single-item measure of positivity and negativity. Cognit. Emot. 23 (3), 453–480. Long, C.R., Averill, J.R., 2003. Solitude: an exploration of benefits of being alone. J. Theory Soc. Behav. 33 (1), 21–44. Lopes, P.N., Brackett, M.A., Nezlek, J.B., Schütz, A., Sellin, I., Salovey, P., 2004. Emotional intelligence and social interaction. Personal. Soc. Psychol. Bull. 30 (8), 1018–1034. Lopes, P.N., Salovey, P., C^ ot�e, S., Beers, M., Petty, R.E., 2005. Emotion regulation abilities and the quality of social interaction. Emotion 5 (1), 113. Marshall, N.J., 1974. Dimensions of privacy preferences. Multivar. Behav. Res. 9 (3), 255–271. McAdams, D.P., 1980. A thematic coding system for the intimacy motive. J. Res. Personal. 14 (4), 413–432. McCormick, I.A., Taylor, A.J., Rivolier, J., Cazes, G., 1985. A psychometric study of stress and coping during the International Biomedical Expedition to the Antarctic (IBEA). J. Hum. Stress 11 (4), 150–156. McGorry, S.Y., 2000. Measurement in a cross-cultural environment: survey translation issues. Qual. Mark. Res. Int. J. 3 (2), 74–81. Nagar, D., Pandey, J., 1987. Affect and performance on cognitive task as a function of crowding and noise. J. Appl. Soc. Psychol. 17 (2), 147–157. Nelson, P.D., Gunderson, E.K.E., 1962. Analysis of Adjustment Dimensions in Small Confined Groups: DTIC Document. Nicolas, M., Bishop, S.L., Weiss, K., Gaudino, M., 2016. Social, occupational, and cultural adaptation during a 12-month wintering in Antarctica. Aero. Med. Hum. Perform. 87 (9), 781–789. Nutt, D., Wilson, S., Paterson, L., 2008. Sleep disorders as core symptoms of depression. Dialogues Clin. Neurosci. 3 (10), 329–336. Palinkas, L.A., 1992. Going to extremes: the cultural context of stress, illness and coping in Antarctica. Soc. Sci. Med. 35 (5), 651–664.
5. Conclusion Despite its limitation of sample size, the present study offers a theoretically-driven model tested using a culturally diverse group of participants. The data represent an agglomerate from five Antarctic stations from different NAP, contrasting with the usual studies con ducted in this field, usually focusing on one station or one programme. The results emphasise the importance of social adjustment to Antarctic stations and the relevance of using measures of social needs in the recruitment process. The study also stresses the central role isolation plays in one’s adjustment and its potential effect on cognitive func tioning, mood and job satisfaction. Declaration of competing interest None. Acknowledgement The authors would like to thank the Polar Research Institute of China, the Alfred Wegener Institute, Antarctica New Zealand, and the Norwegian Polar Institute for granting access to their winter-overs, as well as the participants who gave some of their free time while in Antarctica to make this study possible. References Agho, A.O., Price, J.L., Mueller, C.W., 1992. Discriminant validity of measures of job satisfaction, positive affectivity and negative affectivity. J. Occup. Organ. Psychol. 65 (3), 185–195. Arvey, R.D., Murphy, K.R., 1998. Performance evaluation in work settings. Annu. Rev. Psychol. 49 (1), 141–168. Beaton, D.E., Bombardier, C., Guillemin, F., Ferraz, M.B., 2000. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine 25 (24), 3186–3191. � Bishop, S., Lapierre, J., 2010. Human factors Binsted, K., Kobrick, R.L., Griofa, M.O., research as part of a Mars exploration analogue mission on Devon Island. Planet. Space Sci. 58 (7), 994–1006.
6
C. Jaksic et al.
Polar Science 22 (2019) 100484 Solignac, A., 2004. Influence des moyens de communication sur l’adaptation en situation d’isolement: Le cas d’un groupe d’hivernants en Terre Ad�elie [Influence of communcation channels on adjustment in an isolation context: Case of winter-overs in Ad�elie land]. Universit�e de Reims Champagne-Ardenne. Stewart, W.P., Cole, D.N., 2001. Number of encounters and experience quality in Grand Canyon backcountry: consistently negative and weak relationships. J. Leis. Res. 33 (1), 106. Strange, R., Klein, W., 1973. Emotional and social adjustment of recent US winter-over parties in isolated Antarctic stations. In: Edholm, O.G., Gunderson, E.K.E. (Eds.), Polar Human Biology: Proceedings of the SCAR/IUPS/IUBS Symposium on Human Biology and Medicine in the Antarctic. Heinemann, Chicago, pp. 410–416. Taylor, A.J.W., McCormick, I.A., 1987. Reactions of family partners of Antarctic expeditioners. Polar Rec. 23 (147), 691–700. Thompson, E.R., Phua, F.T., 2012. A brief index of affective job satisfaction. Group Organ. Manag. 37 (3), 275–307. van Winkel, M., Wichers, M., Collip, D., Jacobs, N., Derom, C., Thiery, E., Peeters, F., 2017. Unraveling the role of loneliness in depression: the relationship between daily life experience and behavior. Psychiatry 2 (80), 104–117. Vaughan, G.M., Hogg, M.A., 2014. Social Psychology: French Forest. Pearson Australia, N. S.W. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Personal. Soc. Psychol. 54 (6), 1063. Westin, A.F., 1967. Privacy and Freedom. Atheneum, New York. Westin, A.F., 2003. Social and political dimensions of privacy. J. Soc. Issues 59 (2), 431–453. Winter, D.G., John, O.P., Stewart, A.J., Klohnen, E.C., Duncan, L.E., 1998. Traits and motives: toward an integration of two traditions in personality research. Psychol. Rev. 105 (2), 230. Zeedyk-Ryan, J., Smith, G.F., 1983. The effects of crowding on hostility, anxiety, and desire for social interaction. J. Soc. Psychol. 120 (2), 245–252.
Palinkas, L.A., 2002. On the ice: individual and group adaptation in Antarctica. Retrieved from. http://www.bec.ucla.edu/papers/Palinkas_On_The_Ice.pdf. Palinkas, L.A., Gunderson, E.K.E., Burr, R.G., 1989. Psychophysiological Correlates of Human Adaptation in Antarctica. U.S.A.: Naval Health Research Centre, San Diego, CA. Retrieved from. http://www.dtic.mil/cgi-bin/GetTRDoc?Location¼U2&do c¼GetTRDoc.pdf&AD¼ADA216679. Palinkas, L.A., Gunderson, E.K.E., Holland, A.W., Miller, C., Johnson, J.C., 2000. Predictors of behavior and performance in extreme environments: the Antarctic space analogue program. Aviat. Space Environ. Med. 71 (6), 619–625. Peri, A., Barbarito, M., Barattoni, M., Abraham, A., 2000. The dynamics and the interpersonal and intrapersonal relations within an isolated group in extreme environments. Small Group Res. 31, 251–274. Rona, R.J., Li, L., Gulliford, M.C., Chinn, S., 1998. Disturbed sleep: effects of sociocultural factors and illness. Arch. Dis. Child. 78 (1), 20–25. Rothmann, S., Coetzer, E.P., 2003. The big five personality dimensions and job performance. SA J. Ind. Psychol. 29 (1), 68–74. Roy, D.D., Deb, N., 1999. A Note on Job Satisfaction Level of Scientists and Logistic Personnel in Antarctic Expedition. National Centre for Antarctic and Ocean Research, Department of Ocean Development, New Delhi, India. Ryan, R.M., Deci, E.L., 2000. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am. Psychol. 55 (1), 68–78. Sandal, G.M., van deVijver, F., Smith, N., 2018. Psychological hibernation in Antarctica. Front. Psychol. 9, 2235. Sch€ onbrodt, F.D., Gerstenberg, F.X., 2012. An IRT analysis of motive questionnaires: the unified motive scales. J. Res. Personal. 46 (6), 725–742. Silver, N.C., Dunlap, W.P., 1987. Averaging correlation coefficients: should Fisher’s z transformation be used? J. Appl. Psychol. 72 (1), 146. Smith, S., Haythorn, W.W., 1972. Effects of compatibility, crowding, group size, and leadership seniority on stress, anxiety, hostility, and annoyance in isolated groups. J. Personal. Soc. Psychol. 22 (1), 67.
7