Development of the Moral Disengagement from Refugees Scale: A Psychometric Analysis

674 views

Author/s
Nesrullah Okan1, Muhammed Kızılgeçit2, Tuncay Akıncı3, Gökhan Altuğ4

1 Nesrullah Okan, Firat University, Institute of Education Sciences, Department of Educational Sciences, Elazig, Turkiye.

2 Atatürk University

3 Marmara University

4 Elazığ Provincial Directorate of National Education

Abstract

This study aims to develop a tool that can measure the level of individuals’ moral disengagement from asylum seekers. The study develops the Moral Disengagement from Refugees Scale for this purpose by taking into account the steps for scale development. To do this, the study collected data at three different times for three different purposes (exploratory factor analysis [EFA; n = 342], confirmatory factor analysis [CFA; n = 209], and criterion validity [n = 79]). During the formation of the study group, the participants were informed about the purpose of the study, with sensitivity being shown toward voluntary participation. According to the results from the EFA conducted after obtaining the data, the scale consists of five factors and 28 items, with these five factors explaining 64.251% of the variance. Cronbach’s alpha of reliability for the overall scale was found to be 0.943, with the results for each subscale being within acceptable values. As a result of the CFA, the scale’s subdimensions and items were confirmed, and the goodness-of-fit values are seen to fall within the appropriate reference ranges (χ2 / df= 1.682; RMSEA = 0.066; SRMR = 0.010; IFI = 0.953; CFI = 0.952; GFI = 0.890). The findings obtained as a result of the research reveal the Moral Disengagement from Refugees Scale to be able to be used as a sufficiently valid and reliable measurement tool.

Full Text

Human beings are creatures that harbor and live with many contrasts. Sometimes one can change the values and lifestyle one defends or believes and even consider this to be normal. Bandura (1990) conducted pioneering research on this subject and characterized this issue as moral disengagement he has also conducted numerous studies on this subject. Basically, Bandura (2016) has defined moral disengagement as a series of cognitive mechanism processes that legitimize the individual’s disengagement from moral behavior. In other words, moral engagement can be expressed as cognitively viewing immoral behavior as moral in order to remove the internal unease or distress an individual experiences from immoral behavior. Bandura et al. (1996) discussed moral disengagement under eight components, whose mechanisms are shown in Figure 1. These eight mechanisms are shown in the figure 1.

Figure 1. The mechanisms of moral disengagement (Bandura et al. (1996)).

The situations in which people express moral disengagement are very important. Therefore, many studies have investigated the relationship between moral disengagement and other variables, and the results obtained from these studies are quite interesting. A positive relationship has been found between moral disengagement and various variables such as criminal behavior, antisocial violence, aggression, and substance abuse (Bandura et al., 2001; Bao et al., 2015; Bussey et al., 2015; Gini et al., 2011; Newton et al, 2012; Passini, 2012; Shulman et al., 2011). These research results clearly reveal the link between moral disengagement and many negative variables. In addition, studies on couples’ relationships have also found a positive and significant relationship between moral disengagement and violence (Rollero & De Piccoli, 2020; Rubio-Garay et al., 2019). In fact, moral disengagement can be considered as an individual’s inability to develop empathic skills, with some research results having also shown a negative relationship to exist between moral disengagement and empathy (Paciello et al., 2013; Wang et al., 2017; Zych & Llorent, 2019). As can be understood from these research results, moral disengagement may therefore both distract the individual from empathetic skills, as well as at some point cause the individual to justify themself regarding what is wrong. Undoubtedly, relationships are expected to exist between moral disengagement and many variables. However, the place moral disengagement has regarding individuals’ perspectives toward asylum seekers is undoubtedly very important. People exert great effort to go to countries with a higher level of welfare due to many parameters such as war or economic reasons. These situations carry various risks and can sometimes cost the lives of those who migrate. This is one aspect regarding asylum seekers. Another aspect involves the citizens of the countries where asylum seekers seek asylum. What kind of moral stance the citizens of that country take in terms of asylum seekers is worth investigating. However, no particular measurement tool has yet to be developed that can measure individuals’ levels of moral disengagement toward asylum seekers. The measurement study being conducted here in this context is important in terms of being the first to do so. Based on this importance, the study aims to develop a measurement tool for measuring individuals’ levels of moral disengagement from asylum seekers.

Method

In order to determine the moral disengagement attitudes the citizens from the Republic of Türkiye have toward asylum seekers, the study provides a detailed list of the relevant scale development steps as contained in the literature.

Study Group

The study group of this research consists of individuals between the ages of 18-50 living in different provinces of Türkiye. The purpose of choosing this age range is that these people are considered to be in adulthood. Data were collected at three separate times for the study. Data were first collected from 342 individuals for the exploratory factor analysis (EFA). Data were then collected in the second stage from 209 individuals in order to conduct the confirmatory factor analysis (CFA) regarding the subdimensions and items obtained within the scope of the EFA. Lastly, data were collected from 79 participants and analyzed for criterion validity. While forming the study group, the participants were informed about the purpose of the research, with sensitivity being shown toward voluntary participation.

Scale Development

When examining the literature, no measurement tool is seen to occur that measures individuals’ moral disengagement toward asylum seekers. Based on this gap in the literature, the current study aims to develop a measurement tool for identifying the moral disengagement levels of Turkish citizens toward asylum seekers. The first stage in developing the Moral Disengagement from Refugees Scale, a literature review was conducted and similar studies on this subject were examined. Then, the concept of moral disconnection was explained to individuals at different age levels and half of the 58 individuals were asked to write an essay on this subject and the other half were asked to write ten items about moral disconnection towards asylum seekers. Then, the essay and items were evaluated in the light of the literature. As a result of the evaluations, a pool of 72 items was formed, which were thought to be able to measure individuals’ moral disengagement from asylum seekers. These items were examined by three academicians who’d completed their doctorates in Turkish Language and Literature. After the necessary corrections were made in line with the feedback from the language experts, the items were presented as a single form to 16 experts who’d completed their doctorates in educational psychology, sociology, and social psychology. The experts were asked to evaluate the prepared items according to the criteria on the form (the item should a) remain, b) be corrected, or c) be removed). In addition, they were also asked to suggest corrections or changes to the items.

The form sent regarding the scale items was answered by eight experts. Some items were removed and others were corrected in line with the experts’ opinions. Next, the answers were combined into a single form. The obtained items were evaluated using Lawshe’s (1975) technique. At this stage, the content validity ratio (CVR) for each item were determined using Eq. 1. The CVR is calculated as the ratio of the total number of experts who answered the item should remain to half of the number of experts who gave an opinion on that item minus 1:

                    CVR= [ NG / (  ) ] – 1

where NG refers to the total number of experts who answered that the item should remain, and N refers to the total number of experts who expressed their opinion on the item.

Table 1

As seen in Table 1, the minimum value of the CVRs for eight experts was 0.54 (Veneziano & Hooper, 1997). During the evaluation, each item on the form was examined separately, and a total of 20 items with a CVR < 0.78 were removed from the scale, especially those items with negative (-) or zero (0) CSR values. After the procedures, the remaining 52 items were randomly sorted, and a draft form was created. The participants were asked to fill in this 5-point Likert-type form as follows: 1 = Strongly disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree, 5 = Strongly Agree).

A pilot study was conducted with a group of 54 people to see whether any overlooked expressions or format or spelling errors had occurred on the draft form. After the pilot study, some corrections were made in line with the observations and feedback from the individuals. After these corrections, the instructions were also revised and the scale was finalized. In addition, the personal information form and the Classical Prejudice Scale (Kirişçi-Sarıkaya & Güner, 2021) were included for criterion validity, after which the main application was started. This 52-item scale was sent online to participants from different Turkish provinces in 2022.

Data Analysis

Two package programs were used for analyzing the data obtained within the scope of the scale development. First, the exploratory factor analysis (EFA) was performed, and these analyses were analyzed with the help of the package program SPSS 25. The same program was used for the internal consistency and criterion validity analyses. The package program Amos 24 was used for the Confirmatory factor analysis (CFA) and for performing the necessary analyses.

Findings

This heading presents the psychometric findings that were obtained as a result of the procedures carried out while developing the Moral Disengagement from Refugees Scale (MDRS) and introducing it to the field. This section also tabulates and presents the findings that were obtained as a result of the analyses made by taking into account the procedures followed in the literature for scale development.

Results from the Validity Analyses

For the MDRS that was developed within the scope of the study, EFA was first used to determine what relationships, if any, exist between the items and subfactors used in the study. The EFA considered and examined the Kaiser-Meyer-Olkin (KMO) and Bartlett values. Upon understanding these values to be appropriate, the other procedures of scale development were carried out. The minimum loading value is accepted as 0.30 in the social sciences, with higher loading values indicating higher reliability of the scale (Büyüköztürk, 2015). The analyses conducted in the context of this information eliminated the items with a loading value of 0.50 or less. The KMO and Bartlett’s test values that were obtained as a result of the final analysis are presented in Table 2.

Table 2

As seen in Table 2, the KMO value is 0.907. The literature assumes a KMO value close to 1 to be excellent and less than 0.50 to be unacceptable. Tavşancıl (2010) stated a KMO  0.90 to be excellent, a value of .90 > KMO 0.80 to be very good, a value of 0.70 > KMO  0.60 to be average (mediocre), and a KMO < 0.50 to be bad. As seen in Table 2, the value obtained in this study is accepted as excellent (KMO = 0.907). Likewise, the Bartlett’s test result obtained a value of χ2 = 4,041.515 (p < .000). According to these results, the obtained values are significant, and the data that were used are multivariate and normally distributed.

Table 3

 

Figure 2. The scree plot.

When analyzing Table 3 and the Figure 2, a five-factor structure is seen to emerge when taking the Eigen value closest to 1, and each factor significantly explains the variance. As can be seen in Table 3, the five factors explain 64.251% of the variance. In the social sciences, factors are expected to explain more than 40% of the variance (Büyüköztürk, 2015). When analyzing the common loading values, items with a loading value less than 0.50 were eliminated, ultimately leaving the items that are expressed more strongly on the scale. As a result, 24 items were eliminated from the scale, with the remaining items constituting the final version of the scale.

Table 4

After completing all the procedures, the obtained scale was concluded to consist of five factors and 28 items. Accordingly, the highest loading value of the items on the scale was 0.790, while the lowest loading value of the items was 0.556.

Table 5

As can be seen from Table 5, the scale consists of five factors and 28 items. The final version of the scale has no reverse-scored items. The scores obtained as a result of applying the scale gives both the moral disengagement from refugees score as well as separate scores for each subdimension.

Findings Related to the Reliability Analyses

Running various reliability analyses are recommended when developing a measurement tool. One of the most important reliability analyses is Cronbach’s alpha of internal consistency. Kalaycı (2010) stated the consistency the items in a measurement tool have among themselves to be important when testing the desired structure. In general, a Cronbach’s alpha of internal consistency greater than or equal to .70 is considered acceptable, with internal consistency increasing as the value gets closer to 1 (Büyüköztürk, 2015; Kalaycı (2010); Pedersen & Liu, 2003). Table 6 presents the results from the internal consistency analysis of the MDRS.

Table 6

As can be seen from Table 6, the internal consistency coefficients for the MDRS have been concluded to have the desired values. Cronbach’s alpha of internal consistency for the overall scale and its subdimensions are all greater than .70, with the overall internal consistency having been calculated as 0.943. At the same time, each subdimension achieved a result exceeding the acceptable value.

Confirmatory Factor Analysis

When developing a measurement tool, CFA is used to check whether it meets the fit values accepted in the literature. RMSEA expresses the root mean square error of approximation and SRMR expresses standardized root mean squared residual. In general, RMSEA is expected to have a value less than 0.08, with values closer to zero being considered better. The same is true for SRMR. Meanwhile, the compared fit index (CFI) should be greater than 0.90, with a CFI  0.97 indicating an excellent value (Ekşi et al., 2018).

Figure 3. CFA path diagram of the Refugee Moral Disengagement Scale.

The results obtained as a result of the CFA are shown in Figure 3. Accordingly, all items in the MDRS are understood to fall under five subdimensions and to have significant loading values. This result confirms the previous EFA and shows all the scale items to be able to be considered as elements of a structure. In this context, the 28-item, five-dimensional structure of the Moral Disengagement from Refugees Scale (MDRS) has been tested and confirmed by the CFA. When examining the analyses conducted as part of the CFA, all item loadings are seen to be greater than 0.50. This shows all items on the scale to have sufficient loading values.

Table 7

When examining Table 9, the values obtained from the CFA (Schermelleh-Engel et al., 2003) are seen to generally show good fit (χ2 / df = 1.682; RMSEA = 0.066; SRMR = 0.010; CFI = 0.952; GFI = 0. 890). In light of the obtained results, the data reveal the developed scale to be at an acceptable level.

Correlating the Moral Disengagement from Refugees Scale with Anderson’s (2018) Prejudice Against Asylum Seekers (PAAS) Scale

In order to test the validity and reliability of the MDRS that has been developed for the first time, applying it to similar sample groups with similar scales that have been previously validated and deemed reliable is important. Tavşancıl (2010) recommends the criterion validity test for examining the construct validity of a developed scale.  Therefore, this study uses Anderson’s (2018) Classical Prejudice Scale, which was adapted into Turkish by Okan (2022), to compare it with the similar MDRS in terms of examining the attitudes toward asylum seekers and to test its validity and reliability.

Table 10

Pearson correlation analysis was conducted to determine the relationship between the MDRS as developed within the scope of the research and the PAAS Scale, with the obtained results being shown in Table 10. Due to the PAAS Scale being calculated as a single total score, this total score was taken and included in the analyses alongside the total score and subdimension scores from the MDRS. When examining the analysis results, the PAAS Scale is seen to have a significant relationship with the MDRS (R= 0.809; p < .001). At the same time, all the subdimensions from the MDRS have a significant relationship with the PAAS Scale.

Conclusion and Discussion

When examining the literature, no research is seen to exist examining individuals’ moral disengagement from asylum seekers. Therefore, the need for such a study was considered, and the process was initiated by making the necessary preparations. First of all, the literature was examined in detail, and Bandura (1999) in particular was seen to have made important studies on the concept of moral disengagement and to have even developed a measurement tool in this regard. However, this tool is also understood to lack any dimension regarding asylum seekers. The Moral Disengagement from Refugees Scale (MDRS) was developed here to measure individuals’ moral disengagement from asylum seekers and conducted the necessary analyses. The MDRS is understood to be a reliable and valid measurement tool. Importance is also had in making sure that a developed scale has sufficient validity and reliability values. Büyüköztürk (2015) expressed validity as whether the measurement tool planned for development actually measures the desired construct, while reliability is expressed as the consistency between the answers the participants give. The measurement development processes for the MDRS were followed in stages. Individuals were asked to write both an essay and five items, and 72 candidate items were arrived at in this way. These items were examined by three academicians with doctorates in Turkish Language and Literature. After making the necessary corrections in line with their feedback, the items were then presented as a single form to 16 experts with doctorates in educational psychology, sociology, and social psychology. The experts were asked to evaluate the prepared items according to the criteria in the form as the item should a) remain, b) be corrected, or b) be removed. In addition, they stated being able to make suggestions for corrections or changes regarding the items. Following all these processes, 52 items were identified for and applied to the scale in line with the experts’ opinions. During the common loadings analysis, items with loading values less than 0.50 were eliminated, with the items that were said to express the scale more strongly ultimately remaining. As a result, 24 items had been eliminated from the scale. The remaining 28 items constitute the final version of the scale. According to the EFA results, when taking the Eigen value as 1, a five-factor structure emerged, with each factor significantly explaining the variance. The five factors explain 64.251% of the total variance. In the social sciences, more than 40% of the variance is expected to be explained (Okan & Okan 2021; Scherer et al., 1988). The factor analysis results have concluded the scale to consists of five dimensions and 28 items. CFA was conducted to confirm this result. As a result of the CFA, the five-factor structure of the scale was confirmed, and the fit values were found to be within acceptable limits (χ2 / df = 1.682; RMSEA = 0.066; SRMR = 0.010; IFI = 0.953; CFI = 0.952; GFI = 0.890). In addition, Cronbach’s alpha of internal consistency was examined for the scale and found to be sufficiently reliable. According to all these analyses, the MDRS has sufficient validity and reliability values.

Limitations and Recommendations

No matter how diligent researchers are, any study will have its own set of limitations. This study was limited due to not being able in terms of time and money to reach individuals in every region and province to ask them about their moral disengagement from asylum-seekers. Therefore, the necessary data were collected and analyzed using the population that was accessible. The research attempted to reach every age group if possible and to prevent the above limitation from negatively affecting the research. Due to the research basically being a scale development study, neither parametric nor non-parametric tests were conducted.

According to the results obtained from this study, various suggestions are offered or future studies:

Psycho-trainings can be designed to reduce individuals’ levels of moral disengagement from asylum seekers, and the Moral Disengagement from Refugees Scale can be used for pretest and posttest analyses.

The developed Moral Disengagement from Refugees Scale can be designed for various studies using different variables.


Ethical approval

This study was approved by the Ethics Committee of Kırklareli University (Date: 12.03.2023 No: E-35523585-302.99-68587).

Peer-review

Externally peer-reviewed

Funding

This research received no external funding.

Disclosure statement

The author reports no conflict of interest.

Author’s ORCID numbers

Nesrullah Okan

0000-0002-9496-6417

Figures & Tables

References

Bandura, A. (1990). Selective activation and disengagement of moral control. Journal of Social Issues, 46(1), 27–46.

Bandura, A. (2016). Moral disengagement: How people do harm and live with themselves. Worth Publishers.

Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (1996). Mechanisms of moral disengagement in the exercise of moral agency. Journal of Personality and Social Psychology, 71(2), 364–374. https://doi.org/10.1037/0022-3514.71.2.364

Bandura, A., Caprara, G. V., Barbaranelli, C., Pastorelli, C., & Regalia, C. (2001). Sociocognitive self-regulatory mechanisms governing transgressive behavior. Journal of Personality and Social Psychology, 80(1), 125–135. https://doi.org/10.1037/0022-3514.80.1.125

Bao, Z., Zhang, W., Lai, X., Sun, W., & Wang, Y. (2015). Parental attachment and Chinese adolescents’ delinquency: The mediating role of moral disengagement. Journal of Adolescence44(1), 37–47. https://doi.org/10.1016/j.adolescence.2015.06.002

Bussey, K., Quinn, C., & Dobson, J. (2015). The moderating role of empathic concern and perspective taking on the relationship between moral disengagement and aggression. Merrill-Palmer Quarterly 61(1), 10–29. https://doi.org/10.13110/merrpalmquar1982.61.1.0010

Büyüköztürk, Ş. (2015). Sosyal bilimler için veri analizi el kitabı (21st ed). Pegem Akademi Yayınları.

Ekşi, H., Okan, N., & Mert, H. Ş. (2018). Kişisel Anlam Profili Ölçeği Kısa Formunun geçerlik ve güvenirlik çalışması. Marmara Üniversitesi Atatürk Eğitim Fakültesi Eğitim Bilimleri Dergisi, 47, 157–168. https://doi.org/10.15285/maruaebd.325654

Gini, G., Pozzoli, T., & Hauser, M. D. (2011). Bullies have enhanced moral competence to judge relative to victims, but lack moral compassion. Personality and Individual Differences50(5), 603–608. https://doi.org/10.1016/j.paid.2010.12.002

Kalaycı, Ş. (2010). SPSS uygulamalı çok değişkenli istatistik teknikleri. Asil Yayın Dağıtım.

Kirişçi Sarıkaya, A., & Güner, H. (2021). Sığınmacılara Yönelik Önyargı Ölçeğinin Türkçeye uyarlanması ve psikometrik yönden incelenmesi. KMÜ Sosyal ve Ekonomik Araştırmalar Dergisi, 23(41), 491–504.

Lawshe, C. H. (1975). A quantitative approach to content validity. Personnel psychology, 28, 563–575.

Newton, N. C., Havard, A., & Teesson, M. (2011). The association between moral disengagement, psychological distress, resistive self-regulatory efficacy and alcohol and cannabis use among adolescents in Sydney, Australia. Addiction Research & Theory20(3), 261–269. https://doi.org/10.3109/16066359.2011.614976

Okan, N. & Okan, Y. T. (2021). Kısa dönüşümcü liderlik ölçeğinin Türkçeye uyarlanması ve psikometrik yönden incelenmesi. Turkish Studies-Social, 16(2), 687–700. https://dx.doi.org/10.47356/TurkishStudies.48329

Okan, N. (2022). Adaptation of the Short Attitudes Toward Refugees Scale into Turkish and investigation of its psychometric properties. TRC Journal of Humanitarian Action, 1, 117-125. https://doi.org/10.55280/trcjha.2022.1.3.0013

Paciello, M., Tramontano, C., Nocentini, A., Fida, R., & Menesini, E. (2020). The role of traditional and online moral disengagement on cyberbullying: Do externalising problems make any difference? Computers in Human Behavior103, 190–198. https://doi.org/10.1016/j.chb.2019.09.024

Passini, S. (2012). The delinquency–drug relationship: The influence of social reputation and moral disengagement. Addictive Behaviors, 37(4), 577–579. https://doi.org/10.1016/j.addbeh.2012.01.012

Pedersen, S., Liu, M. (2003). Teachers’ beliefs about issues in the implementation of a student-centered learning environment. Educational Technology Research and Development, 51(2), 57–76.

Rollero, C., & De Piccoli, N. (2020). Myths about Intimate Partner Violence and Moral Disengagement: An Analysis of Sociocultural Dimensions Sustaining Violence against Women. International Journal of Environmental Research and Public Health17(21), 8139. https://doi.org/10.3390/ijerph17218139

Rubio-Garay, F., Carrasco Ortiz, M.Á., García-Rodríguez, B. (2019). Desconexión Moral y Violencia En Las Relaciones de Noviazgo de Adolescentes y Jóvenes: Un Estudio Exploratorio. Revista Argentina De Clinica Psicologica, 28, 22–31. https://doi.org/10.24205/03276716.2019.1089

Scherer, R. F., Luther, D. C., Wiebe, F. A., & Adams, J. S. (1988). Dimensionality of coping: Factor stability using the ways of coping questionnaire. Psychological Reports, 62(3), 763–770.

Schermelleh-Engel, K., Moosbrugger, H., & Müller, H. (2003). Evaluating the fit of structural equation models: Test of significance and descriptive goodness-of-fit measures. Methods of Psychological Research-Online, 8(2), 23–74.

Shulman, E. P., Cauffman, E., Piquero, A. R., & Fagan, J. (2011). Moral disengagement among serious juvenile offenders: A longitudinal study of the relations between morally disengaged attitudes and offending. Developmental Psychology, 47(6), 1619–1632. https://doi.org/10.1037/a0025404

Tavşancıl, E. (2010). Tutumların ölçülmesi ve SPSS ile veri analizi. Nobel Yayın Dağıtım.

Veneziano L., & Hooper J. (1997). A method for quantifying content validity of healthrelated questionnaires. American Journal of Health Behavior, 21(1), 67–70.

Wang, X., Lei, L., Yang, J., Gao, L., & Zhao, F. (2016). Moral Disengagement as Mediator and Moderator of the Relation Between Empathy and Aggression Among Chinese Male Juvenile Delinquents. Child Psychiatry & Human Development48(2), 316–326. https://doi.org/10.1007/s10578-016-0643-6

Zych, I., & Llorent, V. J. (2019). Affective empathy and moral disengagement related to late adolescent bullying perpetration. Ethics & Behavior, 29(7), 547–556. https://doi.org/10.1080/10508422.2018.1521282

We use cookies to help us deliver our services.    More Info