CONSEQUENCES OF PARTICIPATING IN A LONGITUDINAL STUDY OF MARRIAGE
CONSEQUENCES OF PARTICIPATING IN A LONGITUDINAL STUDY OF MARRIAGE
复制标题
参与婚姻纵向研究的后果
DOI：
--
复制
发表时间：
1992
期刊：
影响因子：
--
通讯作者：
E. Douvan
中科院分区：
文献类型：
--
作者：
J. Veroff;S. Hatchett;E. Douvan
使用对黑人和白人新婚夫妇进行的为期四年的研究数据，探讨了调查研究方法，特别是纵向研究中使用的方法，可以影响受访者态度或行为持久变化的可能性。随机选择的初婚夫妇被分配到相对较大的研究组或较小的对照组，前者在研究期间接受更频繁和更深入的采访。一些证据表明，参与专家组程序会对婚姻幸福感产生影响。与对照组相比，研究组在结婚第二年报告的婚姻满意度差异显着较高，反映出更多研究组受访者报告满意度较低的事实。然而，到了第四年，研究组夫妇的婚姻似乎在婚姻质量的几个方面得到了更好的调整。使用调查方法的研究人员想要压制一些问题。也许其中最令人不安的是，他们使用的方法实际上可能会导致他们试图衡量的现象发生短期或长期的变化，换句话说，某些调查研究设计，特别是纵向研究设计，可能包括无意的干预，从而改变态度或行为或两者兼而有之。在这里，我们展示了密歇根大学心理学教授约瑟夫·维罗夫 (JOSEPH VEROFF) 的研究。雪莉·哈切特 (SHIRLEY HATCHETT) 是伊利诺伊大学社会学副教授。伊丽莎白·杜万 (Elizabeth Douvan) 是密歇根大学心理学教授。本文报道的研究得到了国家心理健康研究所 (NIMH) (MH41253) 的资助。我们要感谢与生命历程开发计划成员的合作。 James House 为我们提供了有关本文早期版本的有用建议。民意季刊第 56:315-327 卷（1992 年，美国民意研究协会版权所有。0033-362X/92/5603-0004$02.50 此内容于 2016 年 10 月 12 日星期三 04:29:19 UTC 从 207.46.13.58 下载。所有使用均受http://about.jstor.org/terms 316 Veroff、Hatchett 和 Douvan 在一项针对城市黑人和白人新婚夫妇的婚姻调整和稳定性的为期 4 年的纵向研究中表明，这种影响可能会发生。 Wilson等人（1984）在与这个问题直接相关的研究中发现，让大学生解释他们的约会关系（即告诉他们“列出你能想到的为什么你们的关系……会发展成这样的所有原因”）会对婚姻产生破坏性影响。态度-行为一致性——即与对照组相比，对伴侣的感觉与几个月后是否约会之间的关系，Wilson、Kraft 和 Dunn (1989) 重新分析了数据，发现这种破坏性影响只发生在约会时间较短的夫妻身上。对于我们的研究来说，这些结果不太可能产生关于关系的新材料，而这些材料会破坏彼此的当前态度和未来行为之间的联系。这些结果对我们的研究的意义是，与恋爱关系较短的新婚夫妇相比，人们可能对夫妻在婚前是否同居有类似的期望，从而形成更广泛的经验。威尔逊等人（1989）对约会伴侣之外的其他态度对象进行了平行研究，在这种情况下，他们发现，当受试者被要求对自己的态度做出解释时，一些人会随着时间的推移变得更加积极，一些人会变得更加消极，随着时间的推移，我们预计，与对照组相比，我们对婚姻幸福感的衡量标准会有所增加。此内容下载自 207.46.13.58，星期三，2016 年 10 月 12 日 04:29:19 UTC 所有使用均受 http://about.jstor.org/terms 婚姻研究的影响 317 研究人员还发现，接受采访对政治行为有影响。有一些证据表明，选举前调查提高了一些受访者的实际投票率（Anderson， Silver 和 Abramson 1988；Kraut 和 McConohay 1973；Traugott 和 Katosh 1976）因此，有一些证据表明，我们对初婚夫妇的态度或行为变化可能会产生这种影响。当我们检查主要研究组和对照组中随机抽样的夫妇的婚姻适应和幸福感时，我们可能会发现哪些类型的影响的假设首先，按照 Wilson 等人 (1989) 的研究，我们探讨了第二年研究组的婚姻质量测量是否比对照组有更大的差异。我们认为，正如 Wilson、Kraft 和 Dunn（1989）的研究表明的那样，对于那些认为自己是情侣时间较长的夫妇来说，这两种影响都会更小。
The possibility that survey research methods, particularly those used in longitudinal studies, can effect enduring changes in attitudes or behavior among respondents was explored using data from a 4-year study of black and white newlyweds. Randomly selected couples in first marriages were assigned to either a relatively large study group or a smaller control group, with the former receiving more frequent and more intense interviewing over the study period. Some evidence is found that participating in the panel procedures has consequences for the experience of marital well-being. Compared to the control group, the study group had significantly higher variance in their reported marital satisfaction in the second year of marriage, reflecting the fact that a greater number of study group respondents reported low satisfaction. By the fourth year, however, the marriages of the study group couples appeared to be better adjusted on several dimensions of marital quality. There are some issues that researchers who use survey methodology would like to repress. Perhaps the most disturbing of these is the possibility that the methods they use may actually cause a shortor longterm change in the very phenomenon they are trying to measure-in other words, that certain survey research designs, particularly longitudinal ones, may comprise an unintentional intervention that changes attitudes or behavior or both. Here we present research from an experJOSEPH VEROFF iS professor of psychology at the University of Michigan. SHIRLEY HATCHETT is associate professor of sociology at the University of Illinois. ELIZABETH DOUVAN is professor of psychology at the University of Michigan. The research reported in this article was supported by a grant from the National Institute of Mental Health (NIMH) (MH41253). We would like to acknowledge our collaboration with the members of the Life Course Development Program. James House gave us helpful suggestions about an earlier version of this article. Public Opinion Quarterly Volume 56:315-327 ( 1992 by the American Association for Public Opinion Research All rights reserved. 0033-362X/92/5603-0004$02.50 This content downloaded from 207.46.13.58 on Wed, 12 Oct 2016 04:29:19 UTC All use subject to http://about.jstor.org/terms 316 Veroff, Hatchett, and Douvan imental manipulation in a 4-year longitudinal study of marital adjustment and stability among black and white urban newlyweds that suggests that such effects may occur. We incorporated this experimental design in response to a concern raised by human subjects review boards at both the University of Michigan and the National Institute of Mental Health. Both groups wondered whether long-term, in-depth inquiry into the bases of affection, conflict, difficulties and problems in a marriage, perceptions of each other, attitudes toward gender roles, general levels of wellbeing in the marriage, and the like could raise concerns in a married couple about each other that would not have been considered had we not asked about them. In research directly relevant to the question, Wilson et al. (1984) found that having undergraduates explain their dating relationships (i.e., telling them to "list all the reasons you can think of why your relationship . . . is going the way it is") had a disruptive effect on attitude-behavior consistency-that is, the relationship between feelings toward their partners and whether the couple is dating several months later, as compared to a control group. Since that study, Wilson, Kraft, and Dunn (1989) have reanalyzed the data and have found that the disruptive effect occurs only for couples who had been dating a short amount of time. Various explanations are offered. The ones given greatest credence focus on the assumption that people in longer relationships probably have more consistent schemas about the relationship and hence are less likely to generate new material about the relationship in the interview that would disrupt the connection between present attitude and future behavior. This should be less true for the shorter relationships. The implication of these results for our study is that interviewing newlyweds who had known each other a long time might be less disruptive of their ongoing relationship than interviewing newlyweds who have been in shorter relationships. One might have similar expectations with regard to whether or not the couples lived together before marriage. We might assume that cohabitation would give the couples broader experiences to develop schemas that are more resistant than those of couples who did not live together. Wilson et al. (1989) are doing parallel research on other attitude objects besides dating partners-particularly political figures. In that context, they find that there is a bidirectional attitude change when subjects are asked to generate explanations for their attitudes. Some become more positive over time; some more negative. Following Wilson's lead, we would expect an increased variance on measures of marital well-being over time for couples we intensively interviewed at more points than for a control group interviewed less intensively and for a shorter period of time. This content downloaded from 207.46.13.58 on Wed, 12 Oct 2016 04:29:19 UTC All use subject to http://about.jstor.org/terms Consequences of a Study in Marriage 317 Researchers have also found effects of being interviewed on political behavior. There is some evidence that preelection surveys increase the rate of actual voting of some respondents (Anderson, Silver, and Abramson 1988; Clausen 1968; Kraut and McConohay 1973; Traugott and Katosh 1979; Yalch 1976). Thus, there is some evidence to suggest that the kind of effect that concerned the human subjects board might result from our 4-year prospective study of newlyweds in first marriages. Little is known about attitudinal or behavioral change resulting from data collection. However, we generated two general hypotheses about what types of effects we might find when we examine marital adjustment and well-being among randomly sampled couples in our main study and control groups. First, following Wilson et al.'s (1989) lead, we explored whether there was greater variance in marital quality measures in the second year among the study group compared to the control group. Second, we felt that the general effect of more frequent and intensive interviews would be positive by the fourth year, with the main study group having better marital adjustment and stability than the control group. We felt that both of these effects would be smaller for couples who had considered themselves a couple for a longer period of time, as Wilson, Kraft, and Dunn's (1989) research suggested.