The overall mean attitude of the healthcare students was 3.18 (SD = 0.67), and the means of the sub-scales were as follows: exclusion (M = 2.27; SD = 1.03), similarity (M = 4.44; SD = 1.38), empowerment (M = 3.43; SD = 1.15) and sheltering (M = 3.63; SD = 1.01). Also, the mean score on the GSE was 3.70 (SD = 0.85).
Association among demographics, attitudes and self-efficacy
An independent samples t-test was computed to assess the association between the two-level demographics, attitudes and self-efficacy (Table 1). With respect to the t-tests, differences between participants were found for only two demographics: age and having a relative with ID. To elaborate, there were age-related differences between the participants on all the measures. For the GSE, the younger participants were more efficacious than those who were at least 22 years old t (326) = 3.08, p = 0.002. However, the effect size was very small. Also, for the CLAS-ID, the younger healthcare students were more positive than those who were older, t (326) = 2.45, p = 0.02, with a small effect size. Similar trends were observed on all the sub-scales, with the exception of exclusion, where older students were more in favour of the exclusion of persons with ID than their younger counterparts.
In addition, significant differences were observed regarding having a relative with ID, self-efficacy and attitudes. In terms of self-efficacy, the participants who indicated that they did not have a relative with ID were more efficacious than those who indicated otherwise, t (326) = -4.87, p = 0.001. Although there was no difference between the participants on the CLAS-ID, interesting observations were made on the sub-scales. While those who had a relative with ID were more in favour of more exclusion t (326) = 2.50, p = 0.01, those without a relative with ID indicated more similarities between persons with ID and other members of society, t (326) = -2.92, p = 0.004. Once again, the effect sizes were very small.
A one-way ANOVA was computed to assess the relationship between the three-level demographics, attitudes and self-efficacy of the healthcare students towards persons with ID (see Table 1). Significant differences were observed among three demographic factors: level of study, religion and programme of study. First, differences were observed between the participants on overall attitudes, F (3, 324) = 3.68, p = 0.01, with a small effect size (partial eta squared = 0.03).
Post-hoc comparisons using the Tukey HSD test indicated that healthcare students in their second year were more positive than those in their first year. However, no difference was found between those in their third and fourth years. In terms of the GSE, a significant difference was observed between the participants, F (3, 324) = 12.63, p = 0.001. A post-hoc comparison using the Tukey HSD test found differences between the participants in their third and fourth years only.
Also, differences were observed between the participants in the area of religion. For the CLAS-ID, those healthcare students who identified as traditionalists were more positive than those who identified as Muslim and Christian, F (2, 325) = 5.88, p = 0.003. However, the effect size was very small. A post-hoc comparison using the Tukey HSD test found significant differences between Muslims and traditionalists. Similar observations were made on almost all the sub-scales, with the exception of empowerment where there was significant difference between all the participants. Specifically, participants who identified as traditionalist (a belief in the African religious system) were more in favour of empowerment than those who identified as Christian and Muslim. For the GSE, those who identified as traditionalist recorded higher self-confidence than others, F (2, 325) = 3.55, p = 0.03, with a small effect size. A post-hoc comparison using the Tukey HSD test showed significant differences between the participants.
Although there were no differences between the participants regarding the CLAS-ID, differences were observed for two sub-scales. With respect to exclusion, participants who were allied health professionals were more positive on attitudes than those who were enrolled in healthcare and nursing/midwifery programmes, F (2, 325) = 13.32, p = 0.001. A post-hoc comparison using the Tukey HSD test showed significant differences between healthcare students enrolled in allied health courses and healthcare doctors. However, neither group differed from those enrolled in nursing/midwifery programmes. Additionally, on similarity, those enrolled in nursing/midwifery programme were more positive than those enrolled as trainee doctors and in allied health programmes. A post-hoc comparison using the Tukey HSD test found significant differences between the participants enrolled in allied health programmes and others.
Relationship between self-efficacy and attitude
The relationship between self-efficacy and attitude towards individuals with ID were measured using Pearson’s correlation coefficient. With respect to the causal relationship between attitude and self-efficacy, the bi-directional arrow showed a moderate (r = 0.58) causal relationship between the two latent variables (see figure below for details).
Figure 1 presents the causal relationship between the attitude sub-scales and self-efficacy. However, the model was deemed appropriate as it yielded the following fit indices: chi-square = 4.36, TLI = 0.91, CFI = 0.93, RMSEA = 0.07 and SRMR = 0.05. While very small causal relationships were observed between self-efficacy, exclusion (r = -0.29) and empowerment (r = -0.28), there were moderate relationships between self-efficacy and similarity (r = -0.55) and self-efficacy and shelter (r = 0.32).
Association between attitudes and self-efficacy
Other observations were made between the sub-scales (see Fig. 2). Once again, the fit indices were as follows: chi-square = 4.36, TLI = 0.91, CFI = 0.93, RMSEA = 0.07 and SRMR = 0.05. For example, there was a large correlation between exclusion and empowerment (r = 0.87) and moderate correlations between similarity and shelter (r = -0.430) and exclusion and similarity (r = 0.33).
Relationships between the sub-scales and self-efficacy
Predictors of attitudes
A hierarchical regression was used to assess the importance of self-efficacy in developing healthcare students’ attitudes towards persons with ID in Ghana (see Table 2). In step 1, self-efficacy was regressed directly on attitudes. The results showed that self-efficacy (beta = 0.16, p = 0.01) made a significant contribution, accounting for 21% of the variance in attitudes, F (1, 326) = 8.17, p = 0.005.
In step 2, the addition of the demographic variables contributed to a marginal increase of the variance in attitudes by 0.5%, increasing the total to 21.5%. The overall model contributed significantly to the variance in attitudes, F (8, 319) = 3.27, p = 0.001. In step 2, though self-efficacy (beta = 0.14, p = 0.01) made a significant contribution to the variance in attitudes, age (beta = -0.16, p = 0.01) made the most significant contribution to the variance in attitude.
link