Adaptation of the Short Attitudes Toward Refugees Scale into Turkish and Investigation of its Psychometric Properties

Full Text

 The phenomenon of immigration, which occurs for many various reasons, is a reality that every living creature on Earth faces in one way or another. This reality, which does not belong only to human beings, stands before us as a usual part of the historical process, especially when analyzed in the context of human beings. The Glossary of Immigration Terms (2009) defines immigration as “moving across an international border or within a state.” Be it wars, economic situations, or a wide range of other reasons, humanity has become a part of immigration in one way or another. As a result, a multidimensional communication network emerges between those who migrate, and those, who are migrated to. When we consider the important parameters of this network, it is expected that various conflicts are likely to arise, particularly in terms of culture, religion, language, and understanding. Sam and Berry (2006) stated that individuals experience problems when faced with cultural adaptation, a situation of being primarily stuck between the old and new culture. This, however, only reflects one dimension of the reality of immigration. The examination of the different dimensions and effects of immigration has been of interest to many different disciplines, ranging from psychology to religious studies, and such studies have frequently highlighted the woes of cultural adaptation (Cetrez, 2011). It has been observed that different expressions are used for individuals who are forced to migrate to a different country from their native country. Informative statements regarding these different expressions and definitions are shown in the following figure. 

1. Immigrant: This concept refers to people, who leave their countries for economic reasons and go to other countries legally or illegally.

2. Refugee: This term refers to individuals, who leave their country because they believe that it is not possible for them to survive in their country due to reasons such as persecution, oppression, and torture.

3. Asylum-Seeker: This term refers to a person, who wants to live as a refugee in the country of immigration and who has applied for asylum and is waiting for a response in the form of acceptance or rejection.

4. Temporary Protection: This term refers to persons, who reach the borders of a different country and are granted certain rights as a result of extraordinary events that trigger mass displacement.

Figure 1. Conceptual Framework Formed as a Result of Immigration (Glossary of Immigration Terms, 2009; Turkey and Immigration, 2017). 

Referring to the figure above, although immigration may seem like a single concept, it is a multidimensional and complex phenomenon that leads to the formation of different statuses Berry et al. (2015). Berry et al. noted that an increasing number of asylum seekers are applying for asylum in Western countries as well as in other countries around the world. Such reasons for major drivers of mass immigration can be attributed to the global economic problems that have emerged from the Arab Spring and war situations in Iraq and Syria. The resulting waves of immigration have become an important figure in social sciences and psychology, as well as in political and policy debates (Green-Pedersen & Otjes, 2017; Wagner & Greipl, 2017). Generally, it is the situation of asylum seekers that is taken into account in the framework. In recent times, various studies have been done aiming to assess the attitudes of native citizens of the country of immigration toward new coming asylum seekers (Kotzur et al., 2022). These studies are actually of great importance in terms of keeping the pulse of society on the issue of immigrant individuals. However, the perspective on immigrants and asylum seekers not only has a cultural adaptation dimension, but also diplomatic and political dimensions. In some cases, the phenomenon can lead to significant costs for political authorities. Therefore, the attitudes of the members of society toward immigrants have multidimensional repercussions and different fractions in the eyes of the people. Thus, attitudes of the members of society toward immigrants have multidimensional repercussions and different fractions in the eyes of the people. Attitudes against immigrants have been conceptualized by both classical and contemporary researchers (Allport, 1954; Brown, 2010; Cuddy et al., 2007), and in such conceptualization studies, it was noted that attitudes have cognitive, emotional, and behavioral dimensions. Therefore, it is important to develop the measurement tools in such a way that these three dimensions of individual attitudes are included in order to eliminate various limitations. Kotzur et al. (2022) proceeded with the importance of designing a measurement tool that can measure attitudes toward refugees in the context of all three dimensions. The aim here is to develop a measurement tool that has the validity and reliability to measure the three dimensions of attitudes toward refugees with the minimum number of items. The results of this studies showed that the scale is sufficiently valid and reliable. The adaptation of this scale into the Turkish language was highly desirable, as the adapted version of this scale would be helpful for a country like Turkey—a country that hosts a large number of foreigners, especially refugees, immigrants, asylum seekers, and foreigners under temporary protection status. Therefore, this study aims to adapt the “Short Attitudes Toward Refugees Scale” for the Turkish population. 

Method 

This study is an adaptation study of the “Short Attitudes Toward Refugees Scale” developed by Kotzur et al. (2022) to be implemented in the Turkish population. 

The Pattern of the Study 

A quantitative descriptive screening design serves as the basis for the method of this study. Likewise, studies that developed or adapted talent tests and attitude tests have also been included in the descriptive study group (Büyüköztürk et al., 2019). 

Study Group 

The study group consisted of students enrolled in any undergraduate program at Fırat and Atatürk Universities in Turkey during the Academic Year 2022–2023. Data were collected at different times from the same sample group three times. Data were collected from the first group for exploratory factor analysis (EFA), followed by the second group for confirmatory factor analysis (CFA), and lastly from the third group for criterion validity. Data were collected from a total of 229 students for EFA, and 147 students for CFA. In addition, data were collected from 106 students in total for the criterion validity of the scale. The sample group consisted of individuals between the ages of 17 and 27. 

Translation Works 

Brislin, Lonner & Thorndike (1973) and Cha, Kim, & Erlen, (2007), discussed the translation stages within the framework of the following steps in general. 

1. Translation Into Target Language

2. Evaluation of the Translated Text

3. Retranslation into the Original Language

4. Evaluation of the Retranslated Text

5. Consulting Experts

6. Obtaining the Final Text

7. Application

8. Validity and Reliability

9. Reporting

Figure 2 

Scale Adaptation Steps 

As outlined in Figure No. 2, four experts working as academic staff in the foreign languages departments of different universities were consulted for the translation of the scale from English into Turkish. Furthermore, the German version of the scale was translated into Turkish by two experts. Two different experts from both languages evaluated the resulting translations and they agreed upon the most understandable version. It was critical to take into account the language and cultural structural nuances used in expressions toward refugees. The items that emerged after completing all the stages resulted in the scale being translated back into English and German by experts. The agreed items were then discussed with three academic staff members who specialized in both the language and academic field, and lastly, a final version was agreed upon. 

Data Collection Tools 

The translated form of the Short Attitudes Toward Refugees Scale into Turkish was used, and the Prejudice Scale Toward Asylum Seekers adapted into Turkish by Kirişçi-Sarıkaya & Güner (2021) was used for the criterion validity of the scale. 

Prejudice Against Asylum Seekers Scale 

Anderson’s (2018) Prejudice Scale for Asylum Seekers was adapted into Turkish by Kirişçi-Sarıkaya & Güner (2021). While translating the items of the scale, general rules were respected. The study group of the scale consisted of university students. Data were collected from university students for EFA, CFA, and criterion validity. The results of the EFA showed that the 11-item and one-dimensional structure of the scale explained 47% of the variance. The loading values of the items ranged between 0.792 and 0.489. The one-dimensional structure yielded adequate fit indices as a result of the CFA. In the criterion validity tests, the internal consistency coefficient resulting in a 27% t-test yielded significant and satisfactory values. 

Statistical Tools 

SPSS for Windows 25.0 and AMOS 24.0 package software were used to analyze the statistical data obtained within the scope of the study. Cronbach’s alpha coefficients were analyzed for the internal consistency of the scale. For criterion validity, the Prejudice Against Asylum Seekers Scale was used. Experts’ opinions were taken into consideration for the validity of the content. Both EFA and CFA were used for the structural validity. 

Findings 

The psychometric analysis process of the Short Attitudes Toward Refugees Scale, which was adapted into the Turkish language is discussed in the following section. 

Validity 

A measurement tool is valuable when the validity and reliability analyses are conducted against it. Such analyses enable the measurement tool to achieve a robust scientific objectivity. In this context, the reliability analysis of the measurement tool is presented below. 

Structural Validity 

Both EFA and CFA were used to assess the structural validity of the adapted scale. 

Table 1 

Variance Explained for the Short Attitudes Toward Refugees Scale Based on EFA 

Factor	Explained Variance
Emotional Dimension	36.939%
Cognitive Dimension	33.453%
Behavioral Dimension	20.766%
Total	91.158%

Based on the results of the EFA, there appears to be three dimensions of the Short Attitudes Toward Refugees Scale that explains 91.158% of the variance. The Emotional sub-dimension explains 36.939% of the variance, the Cognitive sub-dimension explains 33.453%, and the Behavioral sub-dimension explains 20.766% of the variance. The obtained results represent an excellent value in the area of social sciences, as it is considered sufficient for the total variance explained to be between 40% and 60% (Çokluk, Şekercioğlu, and Büyüköztürk, 2012). 

Table 2 

KMO and Bartlett’s Test Values 

Kaiser-Meyer-Olkin Sample Adequacy	0.885
Bartlett’s Test of Sphericity	1288.418
	15
	0.000

As displayed in the above table, the KMO value was found to be0.885. The KMO test is important in testing whether the distribution is enough to perform factor analysis as well as to test whether the partial correlations are too small. The KMO value is considered excellent when it approaches 1, and unacceptable when it is below 0.50. According to Tavşancıl (2010), 0.90 and above is considered excellent, 0.80 and above is considered very good, between 0.70 and 0.60 is considered mediocre, and 0.50 and below is considered poor. The KMO value obtained in this study was found to be very good (0.885). However, according to Barlett’s test, the chi-square value was found to be X2 = 1288.418 (p < 0.001). This shows that the values are significant and the used data are multivariate and normally distributed.

Table 3 

Loadings of the Items of the Short Attitudes Toward Refugees Scale 

Items	Cognitive	Emotional	Behavioral
1. Are refugees a threat or an asset for value in Turkey?	0.836
2. Are refugees a threat or an asset for prosperity in Turkey?	0.849
3. How strong is your sympathy for refugees?	0.805
4. How sympathetic are refugees to you?	0.756
I can imagine myself cooperating with refugees.	0.831
I am not against the presence of refugees in the place, where I live.	0.759

Based on the results of the procedures, it can be seen that the scale has a three-factor structure, and the items in this scale have high loading values. It was concluded that the loading values of the cognitive dimension ranged between 0.836 and 0.849, the loading values of the emotional dimension ranged between 0.756 and 0.805, and the loading values of the behavioral dimension ranged between 0.759 and 0.831. The score obtained from–each sub-dimension indicates the total score obtained by the individual from that sub-dimension. A CFA (Confirmatory factor analysis) was then carried out in order to test the structural validity of the scale. Findings related to the CFA are presented below. 

Figure 3. Confirmatory Factor Analysis Path Diagram for the Short Attitudes Toward Refugees Scale 

The three-dimensional and six-item structure of the Short Attitudes Toward Refugees Scale was analyzed through CFA and acceptable fit indices were obtained (X2/sd = 1.88; p > 0.05; RMSEA = 0.062; SRMR = 0.019; NFI = 0.991; CFI = 0.996; GFI= 0.984; IFI = 0.996). The values deemed acceptable by Schermelleh-Engel et al. (2003) and the fit indices of the study are listed in the table below.

Table 4 

The Comparison of Standard Goodness-of-Fit Criteria and Study Results 

Goodness-of-Fit Criteria	Good Fit	Acceptable Fit	Fit Values Obtained Throughout the Study
c2/df	0 ≤ c2/df ≤ 2	2 ≤ c2/df ≤ 3	1.88
RMSEA	0 ≤RMSEA ≤ 0.05	0.05 ≤ RMSEA ≤ 0.08	0.062
SRMR	0 ≤SRMR ≤ 0.05	0.05 ≤ SRMR ≤ 0.10	0.019
NFI	0.95 ≤NFI ≤ 1.00	0.90 ≤ NFI ≤ 0.95	0.991
CFI	0.95 ≤ CFI ≤ 1.00	0.90 ≤CFI ≤ 0.95	0.996
GFI	0.90 ≤ GFI ≤1.00	0.85 ≤ GFI ≤ 0.90	0.984
IFI	0.90 < RFI < 1.00	0.85 < RFI < 0.90	0.996

Criterion Validity of the Scale 

In order to determine the criterion validity of the Short Attitudes Toward Refugees Scale, the sub-dimension titled Classical Prejudice of the Prejudice Scale Toward Asylum Seekers was used and applied to 106 individuals in total. The purpose of using the Classical Prejudice against asylum seekers is that it targets asylum seekers like the adapted scale. The relationship between the sub-dimensions of the adapted scale and Classical Prejudice is shown in the table below. 

Table 5 

The Relationship Between the Scores to be Obtained from the Short Attitudes Toward Refugees Scale and Classical Prejudice Scale 

Variables	1	2	3	4
Classic Prejudice	1	−0.698	−0.703	−0.712
Cognitive	1	0.740	0.773
Emotional	1	0.855
Behavioral	1

**P < 0.001 

Based on the results of the analyses, there is a negative and significant relationship between the subdimensions titled negative Classical prejudice and the subdimensions titled Cognitive subdimension (r = −0.698; p < 0.001), Emotional subdimension (r = −0.703; p < 0.001) and Behavioral subdimension (r = −0.712; p < 0.001). 

Reliability 

Cronbach’s Alpha (α) Internal Consistency Coefficients were used to calculate the reliability of the scale. Findings related to these are presented in this section. 

Table 6 

Internal Consistency Coefficients of the Short Attitudes Toward Refugees Scale 

Scale	Cronbach’s Alpha Value
Cognitive	0.901
Emotional	0.898
Behavioral	0.825
Total	0.940

When we look at the table, the internal consistency coefficient Cronbach’s Alpha value for the Cognitive sub-dimension was α = 0.901, the Emotional sub-dimension was α = 0.898, the Behavioral sub-dimension α = 0.825, and lastly, α = 0.940 for the whole scale. The generally accepted alpha value in social sciences is 0.70 and above (Büyüköztürk, 2010). This shows that the Short Attitudes Toward Refugees Scale adapted into Turkish has suitable value in terms of reliability. 

Discussion and Conclusion 

Both the English and German items of the Short Attitudes Toward Refugees Scale developed by Kotzur et al. (2022) were translated into Turkish by following the translation steps. With the bidirectional translation process and the feedback received from experts, we tried to ensure the understandability of the items as much as possible. Although CFA analysis is considered sufficient enough to assess the structural validity of the scale in the adapted studies, it was also noted that the EFA analysis was calculated in the very same studies. Initially, the EFA was performed and it was determined that the scale had three dimensions. When the results of EFA were analyzed, it was found that the three dimensions explained the variance with a very high value of 91%. This result is considered to be an acceptable percentage in the social sciences, which indicates that the structural validity of the measurement tool is achieved at a high level. Likewise, it is understood that the KMO (0.885) and Barlet’s values (X2 = 1288.418; p < 0.001) also yielded very good results. The item loading values ranged between (0.849–0.746) and were at a high level. Kotzur et al. (2022) reported similar results in their study. 

According to the fit indices obtained from the CFA analyses conducted for this study, generally acceptable results were obtained (X2/sd = 1.88; p < 0.001; RMSEA = 0.065; SRMR = 0.019; NFI = 0.991; CFI = 0.996; GFI = 0.984; IFI = 0.996). In the study conducted by Kotzur et al. (2022), fit indices were found to be (X2/sd = 2,909; p = 0.05; RMSEA = 0.03; SRMR = 0.042; CFI = 0.998). Based on the results of EFA and CFA, it can be seen that the structural validity of the Short Attitudes Toward Refugees Scale, which consists of three dimensions and six items, is confirmed. 

For criterion validity, the Classical Prejudice sub-dimension of the Prejudice Toward Asylum Seekers Scale was used. According to the data obtained from 106 participants, the relationship between stress and the three sub-dimensions of the Short Attitudes Toward Refugees Scale was observed to be negative and significant. The high relation between the Classical Prejudice sub-dimension and all three sub-dimensions indicates that the Turkish version of the Short Attitudes Toward Refugees Scale provides strong criterion validity. 

The internal consistency coefficient was examined in order to determine the reliability of the adapted Short Attitudes Toward Refugees Scale. Cronbach’s alpha value of the three-dimensional scale were found to be at acceptable levels: α = 0.901 for the Cognitive sub-dimension, α = 0.898 for the Emotional sub-dimension, and α= 0.825 for the Behavioral sub-dimension. In this context, when we consider the level of the internal consistency coefficient obtained at the end of our study, it can be said that the reliability of the adapted Short Attitudes Toward Refugees Scale has been achieved. 

In conclusion, based on the results of the EFA, CFA, and Prejudice Toward Asylum Seekers Scale used during validation, it can be said that the adapted Short Attitudes Toward Refugees Scale is a valid and reliable scale. Thus, the Short Attitudes Toward Refugees Scale, which aims to help identify attitudes toward asylum seekers, has strong implications for mass impact in Turkey’s population.

Limitations and Suggestions 

In today’s society, the rise in damaging stereotypes against immigrants may have some respondents be negatively influenced. Such an overall portrait may raise some questions about validity and reliability. In order to minimize the negative repercussions of these instances, the researcher gave briefings of approximately five minutes each to express their general opinions regardless of the current conjuncture. The results obtained from the analysis create a strong conviction that these briefings can be effective. It should also be noted that the current adapted measurement tool would provide researchers with the opportunity to collect data in a short period of time.