I Will Give a Little Help to My Friend – Validation of the German Prosocial Behavior Intention Scale (PBIS-G)
Abstract
Abstract: The present study adapts and investigates the psychometric properties of the German version of the Prosocial Behavioral Intentions Scale (PBIS). The sample included 507 participants. Incongruent with the original study, the PBIS showed low internal consistency (α = .57; ω = .63). The CFA results indicated a good two-dimensional model fit. The PBIS scores showed convergent validity with related instruments. A positive correlation with materialism was observed, so no statements on discriminant validity were made. Additional measurement invariance testing with an existing US sample showed configural measurement invariance, indicating differences in the mode of action in both cultures. Based on these results, the PBIS-G should be used with caution, and further psychometric testing should be conducted to examine discriminant validity and the origin of missing measurement invariance.
The term prosociality summarizes behaviors, such as empathy, sympathy, affection, care, consolation, cooperation, sharing, volunteering, and donation (Hastings et al., 2007; Trommsdorff et al., 2007). Thus, prosocial behavior can be defined as voluntary actions intended to benefit others (Eisenberg, 2006). As a research topic in academia (Mussel et al., 2015), prosocial behavior encompasses a wide range of actions, such as helping, comforting, sharing, and collaborating (Batson & Powell, 2003) as well as recognizing someone's needs or spending time doing tasks for others (Kumru et al., 2004).
Making prosocial behavior and the preceding intentions measurable seems to be of particular importance because of the various positive effects, and Webb and Sheeran (2006) showed that moderate-to-large changes in prosociality can predict small-to-moderate changes in health behaviors. A recent meta-analysis (Hui et al., 2020) found a modest overall mean effect between prosociality and well-being. Prosociality was most related to psychological functioning, showing a modest relationship with psychological malfunctioning and physical health, highlighting the potential benefits of prosociality. In addition, a link was found between participants who engaged in prosocial behaviors and greater meaning in life, as well as increased self-worth (Klein et al., 2015).
On a larger scale, the positive effects of prosocial behavior on an individual's mental and physical health may also indicate a positive effect on an entire community, such as vaccination behavior, where higher prosocial concern is related to higher intentions to vaccinate (Böhm & Betsch, 2022), or environmentally friendly behavior, which can also be classified as prosocial behavior toward the environment as a public good (Otto et al., 2021).
These results prompted the validation of the measurement instruments for prosocial constructs in different languages. For this reason, we validated an existing scale for the assessment of prosociality in German.
Assessing Prosocial Intentions
When it comes to assessing actual prosocial behavior and intentions, the approach often used in the past was to assess behavior through observation. Subsequently, intentions were inferred through observations (e.g., Batson & Darley, 1973; Kahneman et al., 1986; Murphy & Ackermann, 2014). Therefore, prosocial behavior is typically assessed using tasks such as third-party punishment games (e.g., Leliveld et al., 2012) or dictator games (e.g., Ben-Ner & Kramer, 2011; Eckel & Grossman, 1996; Edele et al., 2013; Forsythe et al., 1994). To assess prosociality in a resource-efficient manner, the focus can be laid on the survey of prosocial intentions, which describes the willingness to help others (e.g., Agerström & Björklund, 2009) and, according to Ajzen's (1991)Theory of Planned Behavior, are considered a direct precursor of behavior. To the best of our knowledge, there is currently no generally accepted and valid instrument for assessing prosocial behavioral intentions. It appears that researchers create new items or modify instruments with similar goals in each study (Baumsteiger & Siegel, 2019). In one study, participants were asked to indicate their intentions using a self-developed instrument (Pavey et al., 2011), while in the next study, the same researcher (Pavey et al., 2012) used a self-modified version of the Self-Report Altruism Scale (Rushton et al., 1981). Both instruments show that it is possible to capture prosocial intentions with a small set of self-reports but also reveal the lack of commonly used instruments.
The Prosocial Behavioral Intentions Scale
The Prosocial Behavioral Intentions Scale (PBIS) from Baumsteiger and Siegel (2019) was developed based on the findings of previous work (Pavey et al., 2011, 2012; Penner, 2002; Rushton et al., 1981). The goal was to obtain a brief instrument to assess prosocial behavioral intentions in adults. The instrument is based on two dimensions (helping strangers and helping others), and thus, it is to mention that no model tests were conducted in the original study. With a sample size of N = 247, the PBIS was first found to have sufficient internal consistency (α = .81–.83; Baumsteiger & Siegel, 2019). A positive correlation with prosocial behavior, as assessed by the Self-Reported Altruism Scale (r = .51; Rushton et al., 1981), moral identity assessed using the Moral Identity Scale (r = .50; Aquino & Reed, 2002), and actual prosocial behavior was reported, as individuals who answered optional questions (supportive to the researchers) achieved higher PBIS scores (M = 5.94, SD = 0.96) than those who did not (M = 5.46, SD = 1.21), t(200) = 3.36, p < .001, d = 0.44. The magnitude and direction of the interrelationships between these constructs are consistent with previous studies (Aquino & Reed, 2002; Freund & Blanchard-Fields, 2014; Frimer & Walker, 2009; Winterich et al., 2013). Additionally, binary logistic regression showed that the PBIS was better at predicting prosocial behavior than the Prosocial Intentions Measure (Pavey et al., 2011). In summary, the PBIS provides valid scores for capturing prosocial intentions.
Related Constructs
Analogous to the original scale by Baumsteiger and Siegel (2019), the same instruments will be used, in a translated form, as discriminant and congruent constructs, respectively, for the validation of the German version. Instruments that assess related constructs of prosocial behavioral intentions were found in the Self-Reported Altruism Subscale (SRA; Rushton et al., 1981) and the Moral Identity Scale (MIS; Aquino & Reed, 2002). The SRA was developed to assess the frequency of previous prosocial behaviors. A correlation between current, past, and future prosocial behavior has been demonstrated (Ouellette & Wood, 1998). Individuals with a strong moral identity – people who view doing the right thing as central to their sense of self (Aquino & Reed, 2002; Frimer & Walker, 2009) – are more likely to exhibit prosocial behavior than others (Sage et al., 2006; Winterich et al., 2013). To assess people’s moral identity, the MIS (Aquino & Reed, 2002) has been developed.
Prosocial behavior involves caring for others. Materialism is described as the desire to own expensive items to enhance one's social status (Freund & Blanchard-Fields, 2014). Negative associations have been found between materialism and engaging in prosocial activities (Briggs et al., 2007; Sheldon & Kasser, 1995). The Materialism Scale-Modified (Sirgy et al., 2012) can be used to assess one’s personal materialism. In addition, a relation between prosocial intentions and actual prosocial behavior has been demonstrated (Ajzen, 1991; Smith & McSweeney, 2007).
The Aim of This Study
This study aimed to translate and validate the PBIS (Baumsteiger & Siegel, 2019). This was done with the objective of obtaining a short, uniformly valid instrument for use in several contexts that can be applied to the German-speaking population. For this purpose, the psychometric properties of the PBIS were validated in the German version. Given the large effect in the original study by Baumsteiger and Siegel (2019), we expect the PBIS to correlate positively with SRA and MIS in a large manner and a large negative correlation with MSM.
Method
Participants and Procedure
The survey was conducted online via Unipark. To draw attention to the survey, a call for participation was posted on various social media platforms, such as LinkedIn, and regularly reposted. The final sample included 507 participants. Participants’ ages ranged from 18 to 77 years (M = 36.13, SD = 13.49). There were more women (≈ 65%) than men (≈ 35%) and one person who identified as diverse (< 1%). Most participants came from North Rhine-Westphalia (n = 187, 36.9%), Baden-Wurttemberg (72, 14.2%), and Berlin (58, 11.4%). The majority (61.7%) of participants had a university degree or higher.
To determine a more detailed examination of possible differences between the original study and the results of this study, we carried out further analyses using an existing, English-speaking sample from the United States, with 385 participants (Lorenz, 2021). It consisted of participants between the ages of 18 and 69 years (M = 31.85; 10.36), with 53% women and 47% men. Two participants in the US sample identified as diverse, and one participant refused to answer the question.
Instruments
Translation for all instruments from English was done as follows: First, the authors translated all items into German using gender-neutral language. Afterward, a back-translation was performed by a native German speaker with English language Level C1, equivalent to a TOEFL (Test of English as a Foreign Language) score of 95–120 (see Education First, n.d.). Both translations were compared, and the German translation was adapted, where there was already a large agreement. Finally, the original and finished translations were compared by a native English speaker with a German high school diploma and finally checked for accuracy (original items and German translation are summarized in Table 1).
Prosocial Behavioral Intentions Scale
The PBIS (Baumsteiger & Siegel, 2019) consists of four items using a 7-point rating scale (1 = would definitely not do this to 7 = would definitely do this). It is an instrument with two dimensions, where Items 1 and 3 load on the factor helping friends and Items 2 and 4 on helping strangers. The items asked about willingness to perform a certain action, such as “Comfort someone I know after they experience a hardship,” “Help a stranger find something they lost, like their key or a pet,” “Help care for a sick friend or relative,” and “Assist a stranger with a small task (e.g., help carry groceries, watch their things while they use the restroom).” The scale is thereby instructed as follows: “If you are more likely to complete one task (e.g., helping a stranger find a key) than another (e.g., helping a stranger find a missing pet), please respond to the task you would be more likely to perform.”
Self-Reported Altruism Subscale
The Self-Reported Altruism Scale (SRA; Rushton et al., 1981) was developed to assess the frequency of past prosocial behaviors. Respondents are asked to indicate in 20 items how often they have performed helping behaviors in the past (1 = never to 5 = very often). Items included statements such as “I have helped carry a stranger's belongings (books, parcels, etc.)” and “I have let a neighbor whom I did not know too well borrow an item of some value to me (e.g., a dish, tools, etc.).”
The Self-Importance of Moral Identity Scale
The Moral Identity Scale (MIS) is a self-report instrument consisting of 10 items designed to assess the extent to which a person's self-concept is based on their own moral values, beliefs, and behaviors (Aquino & Reed, 2002). Participants were instructed to think of a person who possesses moral characteristics (e.g., honest, kind), after which agreement or disagreement with certain statements was rated using a 7-point rating scale. Items 1–5 load on the factor Symbolization and Items 6–10 on the factor Internalization, leading to a two-dimensional model, where both factors load on 1 g-factor. The statements include, for example, “I strongly desire to have these characteristics.” Responses ranged from 1 (strongly disagree) to 7 (strongly agree). Higher scores on this instrument indicate a stronger moral identity.
Materialism Scale – Modified
The Sirgy et al. (2012) Materialism Scale-Modified (MSM), consisting of nine items, indicates how important it is for a person to own expensive luxury items for the expression of status and prestige. Here, participants were asked to express their agreement or disagreement with statements such as “Having luxury items is important to a happy life,” whereby these were rated on a 5-point rating scale ranging from 1 (strongly disagree) to 5 (strongly agree). The instrument consisted of three factors loading on 1 g-factor. Items 1–3 load on the factor Materialism-related happiness, Items 4–6 on the factor Materialism-related social recognition, and Items 7–9 on the factor Materialism-related uniqueness.
Further Questions
A brief query on demographic data was added (age, sex, state, educational attainment, and employment status). In addition, we included two attention tests in which participants were asked to select a predetermined response. Finally, to directly assess prosocial behavior, we included two open questions. They were marked as optional questions: “Answering the following two questions is optional, i.e., they can be skipped. If you take the time to answer these questions, you will support the further work by the researchers” with the word optional underlined. However, the real point of interest was whether they answered at all. Explicitly, it was asked, “How would you define a good life?” (Question 1) and “How would you define morality?” (Question 2).
Design
We tested the factor structure of the PBIS using confirmatory factor analysis (CFA). For this analysis, we used the maximum likelihood estimation (MLM)1 with robust SEs and a Satorra–Bentler scaled test statistic. For MLM, a sample size >250 is needed, otherwise, overcorrection of SEs can occur (Yu, 2002).
To assess internal consistency for each instrument, we used McDonald (1999) ω as a measure. In contrast to Cronbach’s α, it does not require an essentially τ-equivalent measurement model with equal factor loadings for all items (Zinbarg et al., 2005). It requires only a τ-congeneric model with variable factor loadings, which is a less strict and more realistic assumption (Jankowsky et al., 2020). We regarded > .70 to be adequately high for the overall factor, whereas the minimum was ω > .30, because the factor saturation of nested factors is considerably smaller (Brunner et al., 2012). Cronbach's α was calculated to represent internal consistency to facilitate comparisons across studies in terms of a popularity measure. Consequently, we analyzed the construct validity of the instruments by Rushton and colleagues (1981), Aquino and Reed (2002), and Sirgy et al. (2012) in their translated versions, and then, the scales were examined for correlation with the PBIS. Finally, participants' PBIS scores were compared with their responses to the optional questions using a t-test. This was done to check whether high PBIS scores directly predicted prosocial behavior. It should be noted that the main focus is on the study of subjective measures and not on the optional questions, since the PBIS asks about intentions that do not necessarily manifest themselves in actual behavior, thus forming an intention–behavior gap (Sheeran & Webb, 2016).
For a comparison between our German and US samples, we first tested the US sample for internal consistency and the model fit identically as described above. Second, we tested the measurement invariance between the already described German sample and the English-speaking sample of people living in the United States, similar to the original validation study (Baumsteiger & Siegel, 2019). We performed the complete data analysis with RStudio (version 4.0.3), using the package Lavaan (Rosseel, 2012).
Results
Model Fit
CFA was conducted to analyze the two-dimensional factor structure proposed in the PBIS (for a graphical representation, see Figure 1). The χ2 test was not statistically significant. Considering the criteria for fit indices, the PBIS showed good fit (Hu & Bentler, 1999; p(χ2) = .674; df = 1; RMSEA < 0.001; CFI = 1.000; TLI = 1.029; SRMR = 0.003). Because the original article did not provide information on the factor structure of the PBIS, we also tested a unidimensional model. The unidimensional model χ2 test was statistically significant, and overall, the PBIS showed a poor fit in all fit indices. The MSM showed an overall good fit, and in contrast, SRA and MIS showed a poor fit, considering the criteria for fit indices (Hu & Bentler, 1999; for detailed information, see Table 2). For the US sample from Lorenz (2021), the χ2 test was not statistically significant. Considering the criteria for fit indices (Hu & Bentler, 1999), the PBIS showed a good fit (p(χ2) = .164; df = 1 ; RMSEA = 0.049; CFI = 0.998; TLI = 0.990; SRMR = 0.007).
Descriptive Analyses
The distributions of the Self-Reported Altruism Scale (M = 58.37, SD = 11.33) and the Moral Intentions Scale (M = 38.28, SD = 8.37) were, according to Shapiro–Wilk testing, normally distributed (p = .17; p = .06). Skewness was 0.20 for the Self-Report and 0.18 for the Moral Intentions Scale, respectively, while kurtosis was 0.06 and −0.10, respectively. The PBIS (M = 23.96, SD = 3.12) and the Materialism Scale–Modified (M = 37.12, SD = 7) were not normally distributed. Their skewness/kurtosis was −1.12; 2.3 for the PBIS and −0.99; 0.63 for the Materialism Scale.
Reliability
The internal consistency for the German version of the PBIS was α = .54, ω = .63. With α = .85, ω = .88, the unidimensional SRA showed high internal consistency, as did the MIS with α = .74 and ω = .83 and the MSM with α = .92, ω = .94. The results of the US version of the PBIS showed high internal consistency, with Cronbach's α = .82 and McDonald's ω = .87. The low α-reliability and questionable ω-reliability of the German version of the PBIS are not in line with our assumptions.
Convergent Validity
The PBIS was positively correlated with past prosocial behavior, r = .35, p < .001, and moral identity, r = .24, p < .001 (see Table 3 for a correlation matrix of the main variables). PBIS scores of people who answered the optional questions were not statistically significantly higher than those who did not answer the optional question (Question 1: t(441.3) = −1.21, p = .226; Question 2: t(502.76) = −1.16, p = .246). The small-to-medium correlations of the convergent scales with the PBIS are not entirely consistent with the expected strong positive correlation. Also not in line with our assumptions is the lack of correlation with the responses to the optional questions.
Discriminant Validity
Unexpectedly, PBIS scores correlated positively with materialism (r = .20; p = .003). No correlation of MSM with SRA or MIS was detected (see Table 3). The results of the discriminant validity test were inconsistent with our assumptions. Also, the convergent hypothesized scales were not significantly negatively correlated with materialism.
Measurement Invariance
A good fit of the two-factor model was established for the German sample's two subsets independently (507 participants) and for the US sample (385). As expected, configural invariance was supported, indicating equality between the factor structures of the two groups. Measurement invariance testing between the US and German samples revealed good model fit for the weak model M2: p(χ2) = .002; df = 4 ; RMSEA = 0.086 ; CFI = 0.982; TLI = 0.946; and SRMR = 0.034 (Hu & Bentler, 1999). However, according to Chen's (2007) recommendations for measurement invariance testing, a deterioration in fit indices of ≤ −.005 in CFI, a change of ≥.010 in RMSEA, or a change of ≥ .025 in SRMR is an indication of noninvariance. This was the case in our analysis, and therefore, only configural invariance can be assumed, indicating a similar factorial structure for the German and US samples (see Table 4 for all results). Oddly, the model for strong invariance did not converge. Due to the small number of items, we could not conduct a further investigation for partial measurement invariance because at least two items per factor must show metric and scalar invariance (Steenkamp & Baumgartner, 1998).
Discussion
General Discussion
The purpose of this study was to translate and validate the PBIS as an instrument to assess prosocial intentions for the German-speaking population. After translation and agreement on the final version of the German PBIS, we conducted CFA to examine the factor structure. Additionally, we examined the convergent and discriminant validity of the instrument. In the original study, no model tests were conducted to confirm the factor structure, and a two-dimensional model was used for this analysis, however, the original PBIS is sometimes also interpreted as a unidimensional instrument (e.g., Anlı, 2019). PBIS exhibited a good model fit in the two-factor solution. However, other studies have identified a unidimensional model as having the best model fit (Anlı, 2019). Analogous to existing studies (Baumsteiger & Siegel, 2019; Ouellette & Wood, 1998), a statistically significant positive correlation was found between achieved PBIS scores and earlier prosocial behavior, but the correlation was considerably lower compared to the original study. Similarly, moral identity (MIS), consistent with existing work, correlated positively with prosocial intentions but was also descriptively lower (Aquino & Reed, 2002; Baumsteiger & Siegel, 2019; Frimer & Walker, 2009). These results are not consistent with either the original study or our assumptions, as a strong correlation was to be assumed by both.
Unexpectedly, a positive correlation between PBIS scores and materialism (MSM) scores was found. This is incongruent with existing studies in which a negative correlation exists between materialism and prosociality (Sirgy et al., 2012) or prosocial intentions (Baumsteiger & Siegel, 2019). Further research on convergent validity should be conducted before the PBIS-G can be used without doubt. However, it is striking that, in our study, past prosocial behavior did not correlate with materialism, which is incongruent with existing research (Sirgy et al., 2012) and our expectations. In addition, the convergent construct of moral identity should theoretically correlate significantly negatively with materialism, which was also not the case in our analysis. Recent research has shown that psychological ownership leads to a boost in self-esteem, which encourages individuals to be more altruistic. Furthermore, this correlation is moderated by materialism (Jami et al., 2021). These results could explain the incongruence observed in the present study. Jami et al. (2021) call into question whether the relation between materialism and prosocial tendencies or intentions assesses discriminant validity. These findings indicate that other constructs need to be used to assess the discriminant validity of a prosociality measure.
People who answered the optional questions did not show statistically significantly higher PBIS scores; this was also incongruent with the original study (Baumsteiger & Siegel, 2019). These results are concerning because, as argued earlier, theoretically, intentions are a direct precursor of behavior according to Ajzen (1991); therefore, there should be a relation between PBIS scores and prosocial behavior. This outcome could also have been caused by methodological weaknesses (see the Limitations and Conclusion section).
The internal consistency coefficients of the German data sets were low and not comparable with those reported by Baumsteiger and Siegel (2019). To determine a more detailed examination of the differences in internal consistency between the original study and the results of this study, further analyses were performed. When analyzing the measurement invariance between the US and German samples, only configural measurement invariance was found. In the US version of the PBIS, a good model fit and high internal consistency were found, congruent with the original validation study (Baumsteiger & Siegel, 2019). These results are in favor of the assumption that the PBIS works differently in the German version than the United States and provide a possible explanation for the low internal consistency in the German version, meaning differences in culture and language could influence reliability outcomes.
This assumption is supported by early research, in which differences in the use of friendly relationships were shown between Germans and Americans. In the United States, informal social resources are being used more in the work context than in Germany (De Graaf & Flap, 1988). Although such results need to be reconfirmed, they give hints on possible differences in social relations explicitly in the term “friend.” A more recently studied topic is the expression “I love you.” Whereas in Germany, the equivalent expression for “I love you” is traditionally reserved for the private disclosure of formal love; in America, “I love you” is professed in a wide range of contexts, including nonromantic relationships (Gareis & Wilkins, 2011), which might indicate that the term “friend” is used differently in these two cultures. Therefore, measurement invariance testing between the original validation samples should be enforced to ensure that no mismatches are made. Altogether, the results of this study provide evidence that the PBIS has a good model fit in both cultures; however, they do not work in the same way. This indicates that both scales may be used to assess prosocial intention, but caution is advised when directly comparing English with German-speaking samples one by one. Generally, caution is advised when using the PBIS-G, since sufficient convergent and discriminant validity could not be conclusively clarified.
Methodological Discussion
In terms of internal consistency, with low Cronbach's α and McDonald's ω values, the PBIS was not convincing. The poor α values could be due to shortness of the scale. Internal consistency increases with more items, which calls into question the general validity of Cronbach's α for very short instruments such as the PBIS (Moosbrugger & Kelava, 2020).
It should also be noted that both measures are homogeneity measures. This leads to the question of whether the PBIS items tend not to be homogeneous for the German-speaking population. This assumption is supported by the low standardized loadings of items PBIS1 and PBIS3 in the German sample (Table 1), which suggests that they are indeed more heterogeneous than in the US sample and is also a possible explanation for the violated weak measurement invariance. The differences, especially in the helping friends factor, can potentially be due to the differences in the term friend already described. The general understanding of what a friend is and what a stranger is seems to differ, which also leads to confusion in a qualitative case study and seems worthy of further investigation (Gareis, 2000). The construct's cultural sensitivity was also supported by the results of the related construct altruism. In the validation of the Facets of Altruism scale, different factor structures were found for the German and the US sample (Windmann et al., 2021).
These differences could also be explained by the fact that prosociality might depend on several features. These features could influence actual prosocial intentions (e.g., one’s own strength, feeling of safety). Prosocial intentions toward a stranger might exist but are discarded out of fear, although not only prosocial intentions toward family or friends exist. Psychosocial entities tend to be clustered sets of real attributes that covary under mutual causation or share underlying common causal mechanisms (Borsboom et al., 2003; Boyd, 1999; Edwards & Bagozzi, 2000). To be theoretically or empirically useful, a test to assess an unobservable entity should capture all the relevant attributes. The nature of psychosocial entities is to be heterogeneous and therefore difficult to analyze. For the benefit of a good model fit and internal consistency, simulation studies may use a too-broad assessed construct and neglect this heterogeneity. Although high levels of reliability are worthwhile, they are not always necessary (Stanley & Edwards, 2016). The results of tests for internal consistency showed that, as already criticized in past research, Cronbach's α is not sufficient as the only criterion for determining the general quality of an instrument (Flake et al., 2017; Flake & Fried, 2020; McNeish, 2018).
An examination of the analyzed instruments SRA and MIS shows that, although the individual values for internal consistency can be rated as good, a different picture emerges when considering the poor model fit of the SRA. This leads to the assumption of a potential trade-off when developing instruments with a sole focus on internal consistency, which in turn leads to broadly used instruments with a poor model fit, with unnoticed weaknesses due to a lack of psychometric analysis or only lack of documentation. In this study, the Materialism Scale–Modified was the only instrument that had an acceptable model fit and high internal consistency.
Limitations and Conclusion
There are some third-party variables that may have influenced the outcome. For instance, people might rate themselves as less likely to “help care for a sick friend or relative,” if they live in isolation or do not have any living friends or relatives, making the PBIS inappropriate for every individual and context. This makes the PBIS only suitable for people with stereotypical western-heterogeneous lifestyles, since there seem to be differences in the types of prosocial behavior between heterogeneous and homogeneous societies according to Baldassarri & Abascal, 2020. Thus, for example, the subscale helping strangers assumes the existing concept of pets and keys (Item-PBIS 2), also assumes social conventions and the need for them, such as taking care of objects when using the toilet (Item-PBIS 2). The items described in the helping friends subscale are also potentially culturally sensitive, as the concern for a relative described in Item-PBIS3 in particular is considered higher and more normal in many cultures (Semnani-Azad et al., 2012).
Furthermore, the responses to the PBIS could be influenced by the participants' desire to seem prosocial in front of themselves or the researchers. The underlined word “optional” was used as an indicator to inform the respondents that the open questions were not mandatory. This raises the possibility that the word optional was not understood and that it would have been better to use a different word, such as voluntary or that the subjects did not realize that the questions were optional and that the task should have been made clearer, e.g., by adding an additional page on which the participants must first agree to make a voluntary extra effort. The assessment would also likely be more effective with an action that requires more time, effort, and thought, which might reveal a clearer distribution of prosociality rates. Nevertheless, further investigation of the relationship between the PBIS-G and the exhibited behavior should be conducted to recommend it for further research.
Therefore, it would be useful to see how well the PBIS predicts types of helping behavior, such as volunteering on a long-term basis or letting a distant acquaintance stay at their house for a longer time. Additionally, it should be considered to add another construct to the model, one from which a theoretical assumption could be made that the correlation should be invariant across cultures (for an example, see Schachler et al., 2019).
If additional studies are conducted on larger samples, attention should be paid to a broader distribution of socioeconomic backgrounds of participants, as individuals in this sample reported high levels of education. Further longitudinal studies are needed to examine the test–retest reliability. In addition, all participants took part in the surveys voluntarily and without any incentives. Therefore, it can be assumed that all participants displayed a certain basic prosociality, which could have influenced the results.
This study illuminates several promising directions for future investigation, although it also presents construct-related questions. For example, the positive correlation with materialism, where elaboration is needed, whether this work represents an outlier, or whether, as studies already indicate, there is a more complex relationship between materialism and prosocial intentions. Still the current version of the PBIS-G should not be used to measure prosocial intentions before additional evidence supporting the instrument's validity has been provided.
References
(2009). Temporal distance and moral concerns: Future morally questionable behavior is perceived as more wrong and evokes stronger prosocial intentions. Basic and Applied Social Psychology, 31(1), 49–59. 10.1080/01973530802659885
(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. 10.1016/0749-5978(91)90020-T
(2019). Adaptation of the prosocial behavioral intentions scale for use with Turkish participants: Assessments of validity and reliability. Current Psychology, 38(4), 950–958. 10.1007/s12144-019-00277-y
(2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1423–1440. 10.1037/0022-3514.83.6.1423
(2020). Diversity and prosocial behavior. Science, 369(6508), 1183–1187. 10.1126/science.abb2432
(1973). From Jerusalem to Jericho: A study of situational and dispositional variables in helping behaviour. Journal of Personality and Social Psychology, 27(1), 100–108. 10.1037/h0034449
(2003). Altruism and prosocial behavior. In I. B. Weiner (Ed.), Handbook of psychology (Part, Vol. 3, pp. 463–484). 10.1002/0471264385.wei0519
(2019). Measuring prosociality: The development of a prosocial behavioral intentions scale. Journal of Personality Assessment, 101(3), 305–314. 10.1080/00223891.2017.1411918
(2011). Personality and altruism in the dictator game: Relationship to giving to kin, collaborators, competitors, and neutrals. Personality and Individual Differences, 51(3), 216–221. 10.1016/j.paid.2010.04.024
(2022). Prosocial vaccination. Current Opinion in Psychology, 43, 307–311. 10.1016/j.copsyc.2021.08.010
(2003). The theoretical status of latent variables. Psychological Review, 110(2), 203–219. 10.1037/0033-295X.110.2.203
(1999). Homeostasis, species, and higher taxa. In R. A. Wilson (Ed.), Species: New interdisciplinary essays (pp. 141–185). MIT Press.
(2007). Beyond just being there: An examination of the impact of attitudes, materialism, and self-esteem on the quality of helping behavior in youth volunteers. Journal of Nonprofit & Public Sector Marketing, 18(2), 27–45. 10.1300/J054v18n02_02
(2012). A tutorial on hierarchically structured constructs. Journal of Personality, 80(4), 796–846. 10.1111/j.1467-6494.2011.00749.x
(2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. 10.1080/10705510701301834
(1988). “With a little help from my friends”: Social resources as an explanation of occupational status and income in West Germany, The Netherlands, and the United States. Social Forces, 67(2), 452–472. 10.1093/sf/67.2.452
(1996). Altruism in anonymous dictator games. Games and Economic Behavior, 16(2), 181–191. 10.1006/game.1996.0081
(2013). Explaining altruistic sharing in the dictator game: The role of affective empathy, cognitive empathy, and justice sensitivity. Learning and Individual Differences, 24, 96–102. 10.1016/j.lindif.2012.12.020
. (n. d.). Englisch Sprachlevel und Bewertungssysteme [
English language levels and rating systems ]. Retrieved from https://www.efset.org/de/english-score/(2000). On the nature and direction of relationships between constructs and measures. Psychological Methods, 5(2), 155. 10.1037/1082-989X.5.2.155
(2006). Prosocial behavior. In G. G. BearK. M. Minke (Eds.), Children's needs III: Development, prevention, and intervention (pp. 313–324). National Association of School Psychologists.
(2020). Measurement schmeasurement: Questionable measurement practices and how to avoid them. Advances in Methods and Practices in Psychological Science, 3(4), 456–465. 10.1177/2515245920952393
(2017). Construct validation in social and personality research: Current practice and recommendations. Social Psychological and Personality Science, 8(4), 370–378. 10.1177/1948550617693063
(1994). Fairness in simple bargaining experiments. Games and Economic Behavior, 6(3), 347–369. 10.1006/game.1994.1021
(2014). Age-related differences in altruism across adulthood: Making personal financial gain versus contributing to the public good. Developmental Psychology, 50(4), 1125–1136. 10.1037/a0034491
(2009). Reconciling the self and morality: An empirical model of moral centrality development. Developmental Psychology, 45(6), 1669–1681. 10.1037/a0017418
(2000). Intercultural friendship: Five case studies of German students in the USA. Journal of Intercultural Studies, 21(1), 67–91. 10.1080/07256860050000803
(2011). Love expression in the United States and Germany. International Journal of Intercultural Relations, 35(3), 307–319. 10.1016/j.ijintrel.2010.06.006
(2007). The socialization of prosocial development. In Handbook of socialization: Theory and research (pp. 638–664). The Guilford Press.
(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. 10.1080/10705519909540118
(2020). Rewards of kindness? A meta-analysis of the link between prosociality and well-being. Psychological Bulletin, 146(12), 1084–1116. 10.1037/bul0000298
(2021). I own, so I help out: How psychological ownership increases prosocial behavior. Journal of Consumer Research, 47(5), 698–715. 10.1093/jcr/ucaa040
(2020). Compiling measurement invariant short scales in cross-cultural personality assessment using ant colony optimization. European Journal of Personality, 34(3), 470–485. 10.1002/per.2260
(1986). Fairness and the assumptions of economics. Journal of Business, 59(4), S285–S300
(2015). It pays to be nice, but not really nice: Asymmetric evaluations of prosociality across seven cultures. Judgment and Decision Making, 10(4), 355–364.
(2004). Olumlu sosyal davranışların ilişkisel, kültürel, bilişsel ve duyuşsal bazı değişkenlerle ilişkisi [Relational, cultural, cognitive, and affective predictors of prosocial behaviors]. Türk Psikoloji Dergis, 19(54), 109–125.
(2012). Punishing and compensating others at your own expense: The role of empathic concern on reactions to distributive injustice. European Journal of Social Psychology, 42(2), 135–140. 10.1002/ejsp.872
(2016). Confirmatory factor analysis with ordinal data: Comparing robust maximum likelihood and diagonally weighted least squares. Behavior Research Methods, 48(3), 936–949. 10.3758/s13428-015-0619-7
(2021). Dataset for: Validation of the revised Compound PsyCap scale (CPC-12R) and its measurement invariance across the US and Germany [Data set]. PsychArchives. 10.23668/PSYCHARCHIVES.5033
(1999). Test theory: A unified treatment. Lawrence Erlbaum.
(2018). Thanks coefficient alpha, we’ll take it from here. Psychological Methods, 23(3), 412. 10.1037/met0000144
(2020). Qualitätsanforderungen an Tests und Fragebogen („Gütekriterien “) [Quality requirements for tests and questionnaires („quality criteria”)]. In H. MoosbruggerA. Kelava (Eds.), Testtheorie und Fragebogenkonstruktion (pp. 13–38). Springer. 10.1007/978-3-662-61532-4_2
(2014). Social value orientation: Theoretical and measurement issues in the study of social preferences. Personality and Social Psychology Review, 18(1), 13–41. 10.1177/1088868313501745
(2015). State-and trait-greed, its impact on risky decision-making and underlying neural mechanisms. Social Neuroscience, 10(2), 126–134. 10.1080/17470919.2014.965340
(2021). The prosocial origin of sustainable behavior: A case study in the ecological domain. Global Environmental Change, 69, Article 102312. 10.1016/j.gloenvcha.2021.102312
(1998). Habit and intention in everyday life: The multiple processes by which past behavior predicts future behavior. Psychological Bulletin, 124(1), 54–74. 10.1037/0033-2909.124.1.54
(2011). Highlighting relatedness promotes prosocial motives and behavior. Personality and Social Psychology Bulletin, 37(7), 905–917. 10.1177/0146167211405994
(2012). “I help because I want to, not because you tell me to” empathy increases autonomously motivated helping. Personality and Social Psychology Bulletin, 38(5), 681–689. 10.1177/0146167211435940
(2002). Dispositional and organizational influences on sustained volunteerism: An interactionist perspective. Journal of Social Issues, 58(3), 447–467. 10.1111/1540-4560.00270
(2012). Lavaan: An R package for structural equation modeling and more. Version 0.5-12 (BETA). Journal of Statistical Software, 48(2), 1–36. 10.18637/jss.v048.i02
(1981). The altruistic personality and the self-report altruism scale. Personality and Individual Differences, 2(4), 293–302. 10.1016/0191-8869(81)90084-2
(2006). Goal orientations and moral identity as predictors of prosocial and antisocial functioning in male association football players. Journal of Sports Sciences, 24(5), 455–466. 10.1080/02640410500244531
(2019). Measuring job crafting across cultures: Lessons learned from comparing a German and an Australian sample. Frontiers in Psychology, 10, Article 991. 10.3389/fpsyg.2019.00991
(2012). Dynamics of helping behavior and cooperation across culture. In Proceedings of international conference on collaboration technologies and systems (CTS) (pp. 525–530). 10.1109/CTS.2012.6261100
(1995). Coherence and congruence: Two aspects of personality integration. Journal of Personality and Social Psychology, 68(3), 531–543. 10.1037/0022-3514.68.3.531
(2012). Linking advertising, materialism, and life satisfaction. Social Indicators Research, 107(1), 79–101. 10.1007/s11205-011-9829-2
(2007). Charitable giving: The effectiveness of a revised theory of planned behaviour model in predicting donating intentions and behaviour. Journal of Community & Applied Social Psychology, 17(5), 363–386. 10.1002/casp.906
(2016). Reliability and model fit. Educational and Psychological Measurement, 76(6), 976–985. 10.1177/0013164416638900
(1998). Assessing measurement invariance in cross-national consumer research. Journal of Consumer Research, 25(1), 78–90. 10.1086/209528
(2007). Sympathy, distress, and prosocial behavior of preschool children in four cultures. International Journal of Behavioral Development, 31(3), 284–293. 10.1177/0165025407076441
(2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132(2), 249–268. 10.1037/0033-2909.132.2.249
(2021). Constructing the facets of altruistic behaviors (FAB) scale. Social Psychology, 52(5), 299–313. 10.1027/1864-9335/a000460
(2013). When moral identity symbolization motivates prosocial behavior: The role of recognition and moral identity internalization. Journal of Applied Psychology, 98(5), 759–770. 10.1037/a0033177
(2016). Identification of confirmatory factor analysis models of different levels of invariance for ordered categorical outcomes. Psychometrika, 81(4), 1014–1045. 10.1007/s11336-016-9506-0
(2002). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous Outcomes (Publication No. 3066425) [Doctoral Dissertation, University of California]. ProQuest Dissertations Publishing.
(2005). Cronbach’s α, Revelle’s β, and McDonald’s ω H: Their relations with each other and two alternative conceptualizations of reliability. Psychometrika, 70(1), 123–133. 10.1007/s11336-003-0974-7
1When considering which estimator to use, either the WLSMV or the MLM, we ultimately decided to use the MLM, because it produces a χ2 distributed statistic if regularity conditions are met (Wu & Estabrook, 2016). Since some used instruments do have Rating Scales of > 5, observed data show a more continuous distribution, whereas WLSMV is specifically designed for ordinal data (Li, 2016).
