Skip to main content icon/video/no-internet
icon backReturn to Entries

Representativeness

Imagine that as a part of a hypothetical study, a life-span researcher finds among a sample of adults aged 55 or above that owning a car is related to both poorer physical health and social isolation. On the surface, these findings may seem counterintuitive or surprising. After all, owning a car during the latter years of life should foster a connection to family, friends, and one’s larger community, in turn leading to better health and a greater sense of social connectedness. Indeed, if these findings do strike the reader as surprising, then likely you are guilty of applying this hypothetical study’s findings to the broader population of adults aged 55 or above. To be fair, reflexively applying these findings to the broader population of adults aged 55 or above is understandable, given that who was actually included in this hypothetical study was not revealed. Indeed, the study’s findings would be less surprising given that the study’s sample was limited to adults aged 55 and above who live in a dense, urban environment, such as New York City, where owning a car is expensive and a potential indicator of an inability or unwillingness to walk or take public transportation.

This hypothetical example illustrates the concept of representativeness as well as its importance when it comes to the implication and interpretation of life-span research. That is, representativeness pertains to who, largely in terms of demographics, is included within or properly captured by the analytic sample in research. Moreover, as the preceding hypothetical example indicates, who is (and is not) represented within a study’s sample helps to frame a study’s implications.

Representativeness Can Vary in Both Quality and Quantity

In one sense, representativeness is qualitative (i.e., some studies have it and some do not). Specifically, depending on the type of sampling strategy, a study may be representative or it may not be. For example, for convenience samples, which entail collecting data in an ad hoc or first-come, first-served basis, it is not really possible to discern who the samples represent. Technically speaking, a given convenience sample (or specifically the demographic composition of a given convenience sample) does represent some larger population, but there is no way really to know with any confidence the true scope or particulars or that larger population. For example, imagine a convenience sample of adults of varying ages who responded to fliers posted at coffee shops and art galleries in downtown Phoenix, Arizona, that promised US$20 to those who filled out a quick survey on health behaviors. What larger population would this sample represent? Put another way, beside those sampled, to whom could we apply or generalize this study’s eventual findings? Unfortunately, the answer is that we could not, at least with any degree of confidence, apply the study’s findings to a broader population or group beyond the specific individuals sampled. Given how important it is to know whom a study represents (as illustrated earlier in this entry) when it comes to evaluating a study and its implications, the fact that it is difficult to discern whom convenience samples actually represent is a significant limitation. In contrast, for probability samples, which incorporate some form of random selection into their sampling procedure, it is straightforward to discern whom they represent. Because, by design, a given probability sample is a random sample of some specified target population, researchers (as well as the consumers of their research) can be confident that any findings generated by that probability sample represent or generalize to that specified target population.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Read next

More like this

Sage Recommends

We found other relevant content for you on other Sage platforms.

close

Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches

close
close
close Signed In to profile You are signed in as:
close
Logged in Institution Institution