Daniël Lakens (2022) discusses the importance of justifying sample sizes in empirical studies. He outlines six approaches to sample size justification: collecting data from the entire population, resource constraints, a-priori power analysis, planning for desired accuracy, heuristics, and no justification. The key is to consider which effect sizes are interesting and how the data informs inferences about them. Researchers should evaluate the smallest effect size of interest, the minimal statistically significant effect, expected effect sizes, confidence intervals, and power for different effect sizes. The value of information depends on how well the sample size aligns with inferential goals. Scientists must balance the cost of data collection against the information gained. The value of information is often non-monotonic, especially with multiple inferential goals. Collecting data from the entire population is straightforward, but when not possible, researchers must consider resource constraints, power analysis, or heuristics. A-priori power analysis helps determine the sample size needed to achieve desired error rates. Researchers should justify effect sizes based on theoretical or practical considerations. Planning for precision involves determining the desired confidence interval width. Heuristics are often based on weak logic and should be critically evaluated. No justification is also a valid approach, but it requires transparency about the lack of justification. The article emphasizes the importance of aligning sample size justification with inferential goals and provides guidelines for researchers to improve their sample size justification.Daniël Lakens (2022) discusses the importance of justifying sample sizes in empirical studies. He outlines six approaches to sample size justification: collecting data from the entire population, resource constraints, a-priori power analysis, planning for desired accuracy, heuristics, and no justification. The key is to consider which effect sizes are interesting and how the data informs inferences about them. Researchers should evaluate the smallest effect size of interest, the minimal statistically significant effect, expected effect sizes, confidence intervals, and power for different effect sizes. The value of information depends on how well the sample size aligns with inferential goals. Scientists must balance the cost of data collection against the information gained. The value of information is often non-monotonic, especially with multiple inferential goals. Collecting data from the entire population is straightforward, but when not possible, researchers must consider resource constraints, power analysis, or heuristics. A-priori power analysis helps determine the sample size needed to achieve desired error rates. Researchers should justify effect sizes based on theoretical or practical considerations. Planning for precision involves determining the desired confidence interval width. Heuristics are often based on weak logic and should be critically evaluated. No justification is also a valid approach, but it requires transparency about the lack of justification. The article emphasizes the importance of aligning sample size justification with inferential goals and provides guidelines for researchers to improve their sample size justification.