手法を比較

	研究におけるサンプリング方法	研究デザインの種類
分野	研究方法論	研究方法論
系統	Process / pipeline	Process / pipeline
提唱年	1950	1963
提唱者	William G. Cochran and Leslie Kish (1950s–1970s)	Donald T. Campbell and Julian Stanley (1963); William Shadish, Thomas Cook, & Donald Campbell (2002)
種類	Framework	Framework
原典	Cochran, W. G. (1977). Sampling Techniques (3rd ed.). John Wiley & Sons.	Campbell, D. T., & Stanley, J. C. (1963). Experimental and Quasi-Experimental Designs for Research. Rand McNally.
別名	sampling strategy, sampling design, probability and non-probability sampling	research designs, experimental and observational designs
関連	3	3
概要	Sampling is the process of selecting a subset of individuals, observations, or units (the sample) from a larger population to study. Sampling methods are broadly classified into probability (random) and non-probability (non-random) approaches. Probability methods—random sampling, stratified sampling, cluster sampling, systematic sampling—enable statistical inference to the population and allow calculation of confidence intervals and margins of error. Non-probability methods—convenience, purposive, snowball, quota sampling—are practical for exploratory or qualitative research but do not support formal statistical generalization. Cochran's Sampling Techniques (1977) and Kish's Survey Sampling (1965) are foundational references; modern applications span surveys, experiments, qualitative studies, and clinical trials.	Research design is the overall structure and strategy of a study, encompassing decisions about how to collect, organize, and analyze data to answer research questions. Major design types include experimental (randomized controlled trials), quasi-experimental (non-random assignment), observational (no manipulation), and qualitative (exploratory, interpretive). Donald T. Campbell and Julian Stanley's 1963 seminal work established systematic terminology for internal validity threats in each design type. Modern classifications (Campbell et al., 2002; Creswell & Plano Clark, 2011) also include mixed-methods designs combining quantitative and qualitative elements.