Differentiate random sampling from random assignment and state why each matters.

Differentiate random sampling from random assignment and state why each matters.

• A. Random sampling eliminates confounds; random assignment increases sample size
• B. Random sampling selects participants from a population for representativeness; random assignment places participants into conditions to ensure equivalence and support causal inference
• C. Random sampling ensures internal validity; random assignment ensures external validity
• D. Random sampling and random assignment are the same process

Answer: (B)

Random sampling is about how you choose who to study from a larger population. By giving everyone in the population an equal chance to be included, the sample tends to reflect that population, which matters for external validity—the degree to which findings generalize beyond the people in your study. Random assignment is about how you place those participants into the different groups or conditions within the study. When assignment is random, the groups are likely to be similar on average before the manipulation, which supports internal validity and makes it possible to infer that any observed differences are caused by the experimental treatment rather than preexisting differences. For example, you might randomly select students from a district to participate so the sample represents that district (external validity). Then you randomly assign those students to use either a new teaching method or the standard method, so differences in test scores can be attributed to the method itself (internal validity and causal inference). Random sampling does not by itself control for confounds, and random assignment does not by itself improve generalizability to a broader population. They serve distinct purposes: sampling for representativeness, assignment for equivalence and causal conclusions.

Random sampling is about how you choose who to study from a larger population. By giving everyone in the population an equal chance to be included, the sample tends to reflect that population, which matters for external validity—the degree to which findings generalize beyond the people in your study.

Random assignment is about how you place those participants into the different groups or conditions within the study. When assignment is random, the groups are likely to be similar on average before the manipulation, which supports internal validity and makes it possible to infer that any observed differences are caused by the experimental treatment rather than preexisting differences.

For example, you might randomly select students from a district to participate so the sample represents that district (external validity). Then you randomly assign those students to use either a new teaching method or the standard method, so differences in test scores can be attributed to the method itself (internal validity and causal inference).

Random sampling does not by itself control for confounds, and random assignment does not by itself improve generalizability to a broader population. They serve distinct purposes: sampling for representativeness, assignment for equivalence and causal conclusions.