relatively large sample sizes (n > 200), classic methods can yield inaccurate confidence intervals and can have relatively poor power. 6 , 7 In practical terms, if Student’s t test fails to reject, one possibility is that there is little or no difference between the groups, or it might be because power is
Rand Wilcox, Travis J. Peterson, and Jill L. McNitt-Gray
Jeffrey Martin and Drew Martin
.e., the p value). All things being equal (e.g., similar measurement error), a study with a large sample size will be more adequately powered than one with a smaller sample size. As a result, research with large samples is more likely to discover small effects that are significant and may be quite
David R. Paul, Matthew Kramer, Kim S. Stote, and David J. Baer
The number of days of data and number of subjects necessary to estimate total physical activity (TPA) and moderate-to-vigorous physical activity (MVPA) requires an understanding of within- and between-subject variances, and the influence of sex, body composition, and age.
Seventy-one adults wore accelerometers for 7-day intervals over 6 consecutive months.
Body fat and sex influenced TPA and MVPA. The sources of subject-related variation for TPA and MVPA were within-subject (48.4% and 54.3%), between-subject (34.3% and 31.8%), and calendar effects (17.3% and 13.9%). Based on within-subject variances, the error associated with estimating TPA and MVPA by collecting 1 to 7 days of data ranged from 28.2% to 13.3% for TPA and 62.0% to 28.6% for MVPA. Based on between-subject variances, detecting a 10% difference between 2 groups at a power of 90% requires approximately 200 and 725 subjects per group for TPA and MVPA, respectively.
Estimates of MVPA are more variable than TPA in overweight adults, therefore more days of data are required to estimate MVPA and larger sample sizes to detect treatment differences for MVPA. Log-transforming data reduces the need for additional days of data collection, thereby improving chances of detecting treatment effects.
The first sport-science-oriented and comprehensive paper on magnitude-based inferences (MBI) was published 10 y ago in the first issue of this journal. While debate continues, MBI is today well established in sport science and in other fields, particularly clinical medicine, where practical/clinical significance often takes priority over statistical significance. In this commentary, some reasons why both academics and sport scientists should abandon null-hypothesis significance testing and embrace MBI are reviewed. Apparent limitations and future areas of research are also discussed. The following arguments are presented: P values and, in turn, study conclusions are sample-size dependent, irrespective of the size of the effect; significance does not inform on magnitude of effects, yet magnitude is what matters the most; MBI allows authors to be honest with their sample size and better acknowledge trivial effects; the examination of magnitudes per se helps provide better research questions; MBI can be applied to assess changes in individuals; MBI improves data visualization; and MBI is supported by spreadsheets freely available on the Internet. Finally, recommendations to define the smallest important effect and improve the presentation of standardized effects are presented.
Andrew C. Cornett and Joel M. Stager
It has been hypothesized that large differences in maximal performance can arise between various geopolitical regions solely on the basis of differing numbers of participants in the target activity. While there is evidence in support of this hypothesis for a measure of intellectual performance, the same relationship has not been examined for a measure of physical performance.
To determine whether the number of participants is a predictor of the best athletic performance in a region.
The 2005–2010 USA Swimming Age Group Detail reports were used to determine the number of competitive swimmers participating in each age group for the 59 local swimming communities in the United States. The USA Swimming performance database provided 50-yd-freestyle times in each community for boys and girls for each age (6–19 y). Simple linear regression was used to examine the relationship between the outcome variable (fastest time) and the predictor variable (log of the number of swimmers) for each combination of age, sex, and calendar year.
The log of the number of swimmers in a region was a significant predictor of the best performance in that region for all 168 combinations of age, sex, and calendar year (P < .05) and explained, on average, 41%, and as much as 62%, of the variance in the fastest time.
These findings have important implications for the development of regional sport strategic policy. Increasing the number of participants in the target activity appears a viable strategy for improving regional performance.
Hotaka Maeda, Chris C. Cho, Young Cho, and Scott J. Strath
, Lauderdale, & Waite, 2017 ). Drops in PA measurement validity can result in loss of power for statistical analyses and unstable study results. In addition, the loss of sample size is equally concerning for loss of statistical power as well as biasing the study sample to compliant participants. Therefore
Yoke Leng Ng, Keith D. Hill, Pazit Levinger, and Elissa Burton
Blood Institute quality assessment tool assessed quality with 14 questions on research objective, study population, participation rate, method of recruitment, sample size, outcome(s), timeframe, levels and assessment of exposure, outcome measures, blinding of assessors, dropout rate, and statistical
Byron Lai, Eunbi Lee, Mayumi Wagatsuma, Georgia Frey, Heidi Stanish, Taeyou Jung, and James H. Rimmer
) targeted dates for systematic searches; (h) the number of included studies; (i) sample size of included studies (mean or range); (j) targeted outcomes, explicit quotes of key results, take home messages, or conclusions; (k) promising elements of effective interventions; (l) research methodological limitations
Laura A. Prieto, Justin A. Haegele, and Luis Columna
identifying major findings; study design characteristics (e.g., sample size, participant characteristics); and dance program characteristics (e.g., location of program, program frequency). The lead researcher used NVIVO software (version 12.0; QSR International, Melbourne, Australia) to code each article for
Chih-Yen Chang and Tsung-Min Hung
methodological improvements. First, the sample size should be increased because only 10 of the 26 studies reviewed recruited more than 20 participants. Small sample sizes can hinder statistical power and the reliability of the observed results. Second, the expert–novice comparison design, the most widely used