학교 성교육 프로그램 효과 메타분석

Abstract

Researches on the effects of sex education programs in the schools of Korea have achieved a quantity growth since the 1980s, reaching the number of 150 or so. They widely vary in terms of outcome variables to test the effects and the resulting effects of sexuality education. But there has been no study to comprehensively consider the effects of the sexuality education program. Thus this study set out to analyze and put together the results of the researches on the effects of sexuality education program in a systematic and objective manner using meta-analysis. Among 148 papers that were searched in the National Assembly Library online, Korea Education and Research Information Service, and Korean Studies Information site, total 54 were selected for analysis. Each of those researches was calculated for effect size by using the formula developed by Hedges & Olkin(1985). After testing homogeneity prior to combining those effect sizes, homogeneity was rejected. After eliminating the outliers and applying the random effect model, weighted average effect sizes were calculated. Another round of homogeneity test was carried out to the groups according to research characteristics. It's also checked if there was statistically significant difference among the groups. The research data were analyzed using the Excel, the meta-analysis program by Lipsey and Wilson(2001) and SPSS program(ver. 12.0). And the criteria of Cohen(1988) were applied to interpret the effect sizes. The research findings were as follows: A. Total 54 papers were analyzed, and their publication dates ranged from 1992 to 2006. Papers for a master's degree were 51(94.4%) and papers published in a journal were 3(5.6%). While 25(46.3%) of the papers studied elementary school students, 21(38.9%) and 8(14.8%) did middle school students and high school students, respectively. The sample sizes of the studies were diverse from minimum 16 to maximum 500. And as for the outcome variables, 50 chose sexual knowledge, 49 sexual attitude, and 2 sexual behavior. B. The sexuality education programs that were applied as an intervention program in those studies were examined. As a result, 15(27.8%) applied a comprehensive sexuality education program and 38(70.4%) applied a program whose contents were not comprehensive. The most heavily focused area of the three was physical and psychological development, which was followed by sexual culture and ethics and understanding of interpersonal relationships in the order. The hours of the sexuality education programs were diverse from 3 to 25; 6(11.3%) had five hours or less, 22(41.5%) from six to nine hours; and 25(47.2%) ten hours or more. While the shortest period of education was two days, the longest one was 28 weeks. C. Since homogeneity was rejected in all the tests, corrected weights according to the random effect model were used to measure the effect size after eliminating the outliers. The overall intervention effects were also tested to see if the effect size was statistically significant. The overall effect sizes of sexual knowledge, sexual attitude, and sexual behavior of all the papers were 1.47(U=2780.87, p=.000), 0.76(U=796.84, p=.000), and 0.07(U=0.66, p=.417), respectively. Sexual knowledge and attitude recorded more than 0.07, which means the effects of sexuality education were great. D. The average effect size was checked according to the research characteristics. As a result, there were no statistically significant differences in sexual knowledge according to all the research characteristics. There were significant differences in sexual attitude according to the category of school, sample size, and hours of education. The middle school students(d=1.74, CI=1.46～2.01) showed the biggest effect size, being followed by the high school students(d=1.25, CI=0.74～1.76) and elementary school students(d=0.72, CI=0.44～1.01) in the order. The effect size was the greatest when the sample size was 20 or less(d=2.08, CI=1.27～2.89), which was followed by 21～50(d=1.44, CI=0.72～2.16) and 51～100(d=1.05, CI=0.70～1.41) in the order. And the effect size was the highest when the hours of education were 10 or more(d=1.44, CI=0.99～1.89), which were followed by 5 hours or less(d=0.91, CI=0.12～1.70) and 6~9 hours(d=0.75, CI=0.42～1.08) in the order. The results indicate that the sexuality education program provided by the schools were very effective in sexual knowledge and attitude. But there were no statistically significant effects found in sexual behavior, which implies that they should focus on the changes to sex behavior in developing a sexuality education program and measuring its effects in future studies on the effects of a sexuality education program. There was a wide variation among the sexuality education programs provided in Korea in terms of the application period and hours of education. It's important to consider the proper education period and hours of education by consulting the sexuality education programs whose effects were proved home and abroad. And 72.2% of the analyzed researches conducted follow-up tests within a week of the end of the program. Since the time of measuring and generating effects can threaten the internal validity of a research, it's critical to take into account the time of generating effects in conducting follow-up tests.