Sampling and generalisability in developmental research: Comparison of random and convenience samples of older adults

Abstract

Researc h in the developmental sciences is based largely on samples of convenience rather than samples drawn at random from the population. The important question of whether results observed in samples of convenience generalise to the larger population has not been studied directly. Because of demographic growth in the proportion of older adults in the population and increases in diversity across the lifespan, it is especially important to address this issue in aging adults. We compared the performance of older adults (65-100 years) on demographic and psychological measures for a random sample of community dwelling adults and two samples of convenience. Signi cant differences were observed on less than half the variables. When differences were present, participants in the convenience samples were advantaged compared to participants from the random sample. Differences were larger in some domains than others but remained small to moderate in magnitude. There were minimal differences in between-person variability and patterns of correlations among variables between the convenience and random samples. Results indicate the need for additional studies contrasting random and convenience samples to explore the parameters of external validity in psychological aging research.

Episodic memory change in late adulthood: Generalizability across samples and performance indices

“…These include careful attention to selecting and documenting the characteristics of the sample (e.g., age, health), the design (e.g., testing interval, sample size) and the measures (e.g., multiple indicators). We are aware of one other recent study that has addressed this issue (see Hultsch, MacDonald, Hunter, Maitland, & Dixon, 2002). Although cross-sectional in design, these authors reported few significant differences when contrasting demographic and psychological performance of older adults from a random sample versus two samples of convenience.…”

“…Although cross-sectional in design, these authors reported few significant differences when contrasting demographic and psychological performance of older adults from a random sample versus two samples of convenience. Specifically, Hultsch et al (2002) observed no differences on word list recall between the random sample and the convenience sample recruited for a study of memory and aging. Further research will examine the extent to which generalizability across samples applies to longer term normal memory aging or to transitional phases of impaired cognitive aging.…”

Growing Up in New Zealand cohort alignment with all New Zealand births

“…Typical approaches to sampling to maximise generalisability of findings in longitudinal studies are use of smaller population frames, such as specific regions or cities, development of structured samples of convenience and selection of an explicitly biased sample. 16 All three approaches were combined into this study's regional recruitment strategy as previously described.…”

Bagaimanakah Persepsi Keterpercayaan Masyarakat terhadap Elit Politik?

“…Kemudian teknik pengumpulan sampel yang digunakan dalam penelitian ini adalah convenience sampling. Pengambilan sampel secara convinience mengarahkan pada pengambilan sampel secara nonrandom (Hultsch, MacDonald, Hunter, Maitland, & Dixon, 2002;Onwuegbuzie & Collins, 2007). Pemilihan metode pengambilan sampel ini dikarenakan pada saat proses penelitian tidak semua sampel mau mengisi kuisioner dan bersedia menjadi kuisioner.…”