(iii) & (iv) There are a number of ways that the researcher could have improved upon the design of the study. First, the population could have been more narrowly defined (e.g., limiting it to adolescents of a given age in a given province). Had this been done, it would have allowed for the use of stronger sampling procedures ensuring greater representativeness. Second, the questionnaire could have utilized a Likert-type scaling system which would have yielded equal-interval data that could have then been analyzed using more powerful statistical methods. This would have yielded more precise data and allowed for a comparative analysis. Third, there could have been greater standardization in getting responses from subjects. This would have contributed to the reliability and validity of data. Had all these things been done, the design would have shifted from non-experimental (see: Ehlers 2003, p. 232) to at least quasi-experimental in nature and would have, therefore, been more internally valid. (iii). No inferential tests were conducted. CAMPBELL, D. T., & Stanley, J. C., 1966. Experimental and quasi-experimental designs for research. Boston. Houghton Mifflin. MELTZOFF, J., 1998, Critical thinking about research. Washington, DC. American Psychological Association. (I) No hypotheses were formulated as the study merely explored whether reasons for avoiding contraception use and related services observed in prior research were the same reasons stopping sample mothers from using contraception/contraception services. No research questions were formulated either, a clear flaw in the study because doing so would have provided more focus and would have allowed for a specific listing of each variable, a listing which, as noted earlier, was absent until the findings were actually reported. (ii) The title of the study was ôAdolescent Mothers Utilization of Contraceptive Services in South Africa.ö This was a somewhat limiting title in that it really did not allu |
