Cross-Cultural Comparison of Economic Behavior Constructs Between U.S. and Kenya Samples

Background and Purpose of Study: The socioeconomic behavior and well-being of poorer households are of great importance to social workers around the world. The study reported here was part of a larger exploratory cross-cultural research that compared social, economic, and psychological outcomes among samples of poorer households from the Republic of Kenya and the U.S. during the recent deep recession. The purpose of this study was to assess economic behavior constructs between the two samples. Specifically, the study answered the following research question: Are there measurement- or structural-level differences in financial capability (FC), economic strain (ES), and purchasing behavior (PB) across the groups?

Methods: The samples for this study are from populations living in publicly-subsidized housing in a mid-sized city in Southeastern U.S. and in the Eastern part of Nairobi, Kenya between 2009 and 2010. Participants in both groups self-identified as “heads of households.” Non-probability samples were used and data collected via face-to-face interviews and self-administered questionnaires. FC is defined as the ability that comes from knowledge and experience that lower income households use to meet their economic needs by acting in their best economic interests. The items were selected from the NC 1013 Economic and Psychological Determinants of Household Savings Behavior questionnaire. ES was developed by Pearlin et al. (1981) and items refer to the subjective feelings of distress or pressure associated with a perception of inadequate financial resources to provide for the needs of family.  PB, from the NC 1013 questionnaire, capture the impulse to make purchases or buy items without regard to the short- or long-term economic consequences of the action. Multiple group structural equation modeling was used for analysis. Items in all three constructs were measured on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). A two-group structural equation modeling with three constructs using Mplus 7.0 was conducted.

Results: Measurement invariance was established, χ² (198, 383) =287.80, p <.001, RMSEA=.049, (90% CI: .036-.061), NNFI=.951, CFI=.954, indicating that the same constructs were being measured across cultures. Homogeneity of variances and covariances was not established, however, equality of correlations was established. The groups significantly differed by latent means, Δχ²_SB (1) = 4.85, p < .05 on the PB construct, with a small effect size (d = .27). This finding indicates that PB was significantly lower in the U.S. group (M = 2.50, SD = .48) than in the Kenya group (M = 2.67, SD = .66). There were no mean differences in FC and ES between the groups.   

Conclusions and Implications: Though limited in external validity, results establish the potential for studying socio-economic constructs internationally with very improbable groups. Programs aimed at enhancing the financial capability of lower income households are critically needed across the world. Given the negative wording of the PB construct, poorer households in the Kenya sample were in fact savvier in how they spent monies than their U.S. counterparts. PB is very important in enhancing FC among poorer households. The results of this study indicate a further need for increased financial education among lower-income households around the world.