Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

The Importance of Samples

Although we don’t ignore biological differences (see Berenbaum et al. 2011; Muehlenhard and Peterson 2011), the mission of Sex Roles is to better understand “the underlying processes and consequences of gender role socialization, gendered perceptions and behaviors, and gender stereotypes…” (, para. 1). Since these all derive from the underlying cultural beliefs, it is essential that any papers published in Sex Roles carefully consider what these underlying beliefs are and how they might relate to the variables being investigated in the reported research. Theory about the cultural beliefs comes from published research based on samples in different countries. In building theory and reviewing relevant empirical studies, we ask that our authors be sensitive to and note what types of samples were used and where the studies were done. Authors need to consider whether the data from a cited study are directly relevant to their own sample, if the cited study uses a very different type of sample or comes from another country.

In order to analyze the societal or cultural beliefs of a group, one must be able to clearly define that group in order to understand the relevant social context. At the simplest level, we can assume that individuals in a country share a set of cultural beliefs. A sample group needs to be based in a single country. We ask our authors not to generalize about “Western” or “Eastern” culture since these groups are not clearly defined. But, a further issue is that the individual countries may have quite different beliefs from each other, even if in a region such as Europe. Well defined groups within a country may also have their own ideas about gender. Thus, a study may need to develop an analysis for each ethnic group in the sample. Age must also be considered. There is empirical evidence that gender beliefs in the U.S. have changed over time (Twenge 1997). Thus, the authors may have to form meaningful age groups in developing theory about how gender beliefs impact the group. Life stage is related to age, but is a separate consideration. Within a country or even in particular region, university students may be very different from the general public, even if their ages are similar. In (Frieze et al. 1991), we discuss how ignoring the social context may have masked important gender differences in research on achievement.

These concerns are reflected in this journal in various ways. First, we ask that the sample be carefully explained in the Sample section of the Method. Second, each group needs to be a relatively homogeneous sample. Given the focus of the journal, one group is always gender of the study participant. Women and men in the sample need to be considered separately, since their gender beliefs are quite likely to differ (McHugh and Frieze 1997). If the sample is undergraduate students, much older students should be omitted from the sample so that we are able to make generalizations about young undergraduate women and men. Generally, anyone aged 30 or older is likely to have been socialized at a very different time than younger students and may have had different life experiences. It is especially important that any sample demographics be comparable for the women and men or girls and boys in the sample, given the focus on gender for this journal. It is quite possible that if age or some other relevant variable is differentially represented in the samples of women and men, what appears to be a significant gender difference may indeed be an age effect.

Mechanical Turk ( is a recently developed system that allows researchers to enter a survey on-line. Participants volunteer and receive a small payment for completing the survey. MTurk samples present a number of potential questions in terms of samples. The very fact of their participation in this procedure makes the women and men in the MTurk sample different from the general public. We ask that this be considered in understanding data generated from this type of sample. We also require that at least the country in which they live be identified, and that sample groups not be formed across countries. Since age is likely to vary greatly within these samples, the author needs to form age groups or control for age statistically. Any other major demographic factors such as education level may also relate to the variables of the study. As much information about the characteristics of the sample as possible should be collected so that these variables can be considered in analyzing the data.


  1. Berenbaum, S. A., Blakemore, J. E. O., & Beltz, A. M. (2011). A role for biology in gender related behavior. Sex Roles, 64, 804–825. doi:10.1007/s11199-011-9990-8.

  2. Frieze, I. H., Sales, E., & Smith, C. (1991). Considering the social context in gender research: The impact of college students’ life stage. Psychology of Women Quarterly, 15, 371–392. doi:10.1111/j.1471-6402.1991.tb00414.x.

  3. McHugh, M. C., & Frieze, I. H. (1997). The measurement of gender-role attitudes: A review and commentary. Psychology of Women Quarterly, 21, 1–16. doi:10.1111/j.1471-6402.1997.tb00097.x.

  4. Muehlenhard, C. L., & Peterson, Z. D. (2011). Distinguishing between sex and gender: History, current conceptualizations, and implications. Sex Roles, 64, 791–803. doi:10.1007/s11199-011-9932-5.

  5. Twenge, J. M. (1997). Attitudes toward women, 1970–1995: A meta-analysis. Psychology of Women Quarterly, 21, 35–51. doi:10.1111/j.1471-6402.1997.tb00099.x.

Download references

Author information

Correspondence to Irene H. Frieze.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Frieze, I.H. The Importance of Samples. Sex Roles 70, 357–358 (2014).

Download citation