EXPERIMENTAL AND QUASI-EXPERIMENTAL STUIDES An extensive review of the premarital education literature found a total of 12 studies which included control or comparison groups against which the effects of the intervention was evaluated, and therefore can be categorized as being experimental in design (Bader et al., 1980; Bagarozzi et al., 1984; Boike, 1977; D’Augelli et al., 1974; Hahlweg et al., 1998; Markman, et al., 1993; Miller et al., 1976; Parish, 1992; Pretorius et al., 1992; Ridley et al., 1981; Schlein, 1971; Stanely, et al., 2001). As can be seen in Table 1, all but two of these studies (Hahlweg et al., 1998; Parish, 1992) used random assignment to insure equivalence of experimental and control groups. The two studies which did not use randomization procedures are separated out and identified as “quasi-experimental” in design in Table 1. Sample Characteristics Of the 12 experimental and quasi-experimental studies reviewed, two early studies (D’Augelli et al., 1974; Schlein, 1971) did not give a detailed report of how they obtained their samples. For the most part, participants of the remaining studies were recruited through three primary sources: community sources (media, clinics, etc.), religious organizations, and college courses. All of the studies focused on couples marrying for the first time, but the relationship status of the couples participating was varied and somewhat ambiguous in that 8 of the 12 studies identified their couples as “engaged” or “couples planning to marry” (Bader et al., 1980; Boike, 1977; Hahlweg et al., 1998; Markman, et al., 1993; Miller et al., 1976; Parish, 1992; Pretorius et al., 1992; Stanley et al., 2001), two studies identified them as “premarital couples” (Bagarozzi et al., 1984; Ridley et al., 1981), and two studies identified their participants as “dating couples” (D’Augelli et al., 1974; Schlein, 1971). Nine of the 12 studies were conducted in the United States, while the other 3 were conducted in Canada (Bader et al., 1980), South Africa (Pretorius et al., 1992) and Germany (Hahlweg et al., 1998), respectively. Only 6 of the 10 experimental studies done in the United States reported detailed information about the demographic characteristics of their samples. Of those providing information, all reported that their samples were almost exclusively made up of Euro-American, middle-class couples. Similarly, the study done by Pretorius and colleagues (1992) in South Africa only included a sample of “white South Africans” and the study done by Hahlweg and colleagues (1998) in Germany also reported a homogeneous sample in terms of ethnic diversity and social-economic status. It is clear that experimental studies investigating the effectiveness of premarital education have been done with a very homogeneous population with regards to ethnic and SES diversity. Due to the lack of diversity in the populations being reached by such research, caution should be used in generalizing their results to other populations. Sample size. Of the 12 experimental and quasi-experimental studies reviewed, the size of the samples utilized for experimental and control groups varied widely (range n = 7 to 139). Of the 25 samples reported (i.e., 13 experimental groups & 12 control groups), 48% of them can be categorized as being small in size (e.g., n = 1-29); 36% as being moderate in size (e.g., n = 30-59), and the remaining 16% as being large in size (e.g., n = 60+). It is important to note that these are the reported sample sizes for when the studies began. Due to attrition and other factors, some of the studies actually had smaller samples in their post-test and follow-up evaluations. 
đang được dịch, vui lòng đợi..
