The use of categorical variables in regression involves the application of coding methods. The purpose of this paper is to describe how categorical independent variables can be incorporated into regression by virtue of two coding methods: dummy and effect coding. The paper discusses the uses, interpretations, and underlying assumptions of each method. In general, overall results of the regression are unaffected by the methods used for coding the categorical independent variables. In any of the methods, the analysis tests whether group membership is related to the dependent variables. Both methods yield identical R2 and F. However, the interpretations of the intercept and regression coefficients depend on what coding method has been applied and whether the groups have equal sample sizes.

A short version of Buri's (1991) 30-item Parental Authority Questionnaire (PAQ), a widely used measure of Baumrind's (1971) model of authoritative, authoritarian, and permissive parenting styles, has been developed in this study. The participants were 3,025 middle and high school students from Oman. The sample was randomly divided into 2 subsamples. The first sample (N = 1,504) was used for the development of the short version, and the second sample (N = 1,521) served as the validation sample. The results indicate that a reduced 20-item PAQ fit the data better than the 30-item PAQ. The short version evidenced adequate validity and internal consistency. It was discussed and confirmed that the short version of the PAQ utilizes Baumrind's model of parenting styles as well as the original long version of the PAQ, and functions appropriately. This result is consistent with previous research.

The current study investigated the associations between three maternal and paternal parenting styles, moral intelligence, academic self-efficacy and learning motivation in three serial mediation models. Omani adolescents enrolled in 7th to 11th grades (N = 296) responded to an online survey containing demographic items and scales measuring the variables noted above. Results of Path Analysis indicated that the three models had a good overall fit. In detail, the three paternal styles (authoritative, authoritarian and permissive) had direct associations with moral intelligence and indirect associations with learning motivation. However, only two maternal parenting styles (i.e., authoritative and authoritarian) correlated directly with learning motivation and these two styles did not associate with moral intelligence. All effects were in the hypothesized direction except the effect of authoritative maternal and paternal styles. Moral intelligence had a positive direct correlation with students’ academic self-efficacy and learning motivation. Moral intelligence also mediated the negative associations between three types of fathers’ parenting styles and students’ motivation. Academic self-efficacy had a positive association with students’ motivation. These findings provided useful insights about the various association between external factors (e.g., parenting styles), internal factors (i.e., moral intelligence and self-efficacy) and students’ motivation among adolescents in middle and high schools.

This study examined the effects of classroom assessment practices on students' achievement goals. The study included 1,636 ninth grade students and 83 science teachers from Muscat public schools in Oman. Results from hierarchical linear modeling techniques showed that class contextual features and teachers' teaching experiences and assessment practices interacted significantly with students' characteristics in influencing students' achievement goals. Recommendations, implications, and suggestions for practice and future research are discussed.