Sampling issues in parent-child nursing research: implications for nursing practice.

Abstract

A S NURSES ARE increasingly expected to read and incorporate research findings into their practice, emphasis on the evaluation of such research becomes critical. Several critical and integrative reviews of nursing research have been published over the past decade (Beal & Betz, in press; Betz & Beal, in press; Brown, Kelen, & Moser, 1985; Cronenwett, 1982; Diers & Molde, 1979; Jacobsen & Meininger, 1985; Moody et al., 1988; Perlmutter, 1985; Selby et al., 1990; Silva, 1986; Strehle, 1981; Uman & Urman, 1990; Werley & Fitzpatrick, 1984, 1985, 1986). Only a few reviews have evaluated the research published in the subspecialty of parentchild health nursing (Beal & Betz, in press; Betz & Beal, in press) or specifically the sampling methodology used (Selby et al., 1990). Because the sampling method used during the conduct of the research so directly impacts on the reader’s ability to use the research findings in clinical practice, we chose to evaluate the sampling methods used in the parentchild research published during the past decade. The purpose of this article is to describe our review and the implications of our findings for the clinical practice of parent-child health nurses. Few studies have scientifically reviewed sampling methodology. Moody and colleagues (1988) in a review of the nursing research published from 1977 to 1988, found that most of these studies reported using small nonrandomized samples. Also, Jacobsen and Meininger (1985) reported that sample sizes were small and that sampling methods were not clearly described. Selby et al. (1990) in a more recent review of 30 randomly selected research reports published in 1986, found that nearly 97% contained at least one major deficiency in sampling methodology. Some of the deficiencies were as follows: (a) failure to describe the sampling frame, sample size, number of refusals, withdrawals and /or cases lost, and sampling methods used; (b) inappro-