PK}j7z4VVrefs.MYD?Hopkins, W G20065Estimating sample size for magnitude-based inferences63-67Sportscience10Qconfidence limits, research design, statistical power, Type 1 error, Type 2 error0Sample-size estimation based on the traditional method of statistical significance is not appropriate for a study designed to make an inference about real-world significance, which requires interpretation of magnitude of an outcome. I present here a spreadsheet using two new methods for estimating sample size for such studies, based on acceptable uncertainty defined either by the width of the confidence interval or by error rates for a clinical or practical decision arising from the study. The spreadsheet includes a section for estimating sample size by the traditional method, which requires sample sizes three times greater than those provided by the new methods. The key issues and statistical principles underlying sample-size estimation are outlined in an accompanying slideshow and conference poster. "http://sportsci.org/2006/wghss.htm]Sport and Recreation, AUT University, Auckland 0627, New Zealand. Email: will=AT=clear.net.nz?BGuyatt, G.
Jaeschke, R.
Heddle, N.
Cook, D.
Shannon, H.
Walter, S.19950Interpreting study results: confidence intervals169-73$Canadian Medical Association Journal152?#Hopkins, W G
Hawley, J A
Burke, L M1999@Design and analysis of research on sport performance enhancement472-485+Medicine and Science in Sports and Exercise31perf, stats, relyreviewjour?&Petersen, C J
Wilson, B D
Hopkins, W G2004AEffects of modified-implement training on fast bowling in cricket 1035-1039Journal of Sports Sciences22?Hopkins, W G20049Bias in Bland-Altman but not regression validity analyses42-46Sportscience8^calibration, method comparison, random error, systematic error, standard error of the estimateAn instrument that has been calibrated against a criterion measure with a sample of subjects is sometimes checked against the criterion in a validity study with another sample. In a spreadsheet-based simulation of such calibration and validity studies, a Bland-Altman plot of difference vs mean values for the instrument and criterion shows a systematic proportional bias in the instrument's readings, even though none is present. In contrast, a regression analysis of the criterion vs the instrument shows no bias. The regression analysis also provides complete statistics for recalibrating the instrument, if bias develops or if random error changes since the last calibration. The Bland-Altman analysis of validity should therefore be abandoned in favor of regression.'http://sportsci.org/jour/04/wghbias.htmSport and Recreation, Faculty of Health, Auckland University of Technology, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?Hopkins, W G2006[Viagra at altitude and other performance-related highlights of the ACSM 2006 annual meeting1-7Sportscience10:elite athletes, ergogenic aids, nutrition, tests, trainingcAcute Strategies: Effects of ibuprofen, pre-tensing, pre-cooling, circadian rhythym, warm-ups, ice-cold recovery, bike fitting, stretching, post-activation potentiation, whole-body vibration, and massage. Altitude: Viagra for performance at high altitude; the live-high train-low mechanism debate; effects of adaptation to artificial altitude. Mechanisms: maximum effort is related to a critical fatigue level in muscle; hard training stimulates EPO; evidence that muscle pH limits intense exercise. Nutrition: training on low carbohydrate; galactose vs other carbs and caffeine; milk protein, amino acids and cherry juice for recovery; colostrum for training; vitamin C and fish oil for asthmatics; echinacea stimulates EPO; caffeine for team-sport and tennis performance; effects of mild hypohydration. Performance Genes: minor findings. Tests and Technology: using modeling to optimize cycling performance; soccer tests; mountain-bike suspensions. Training: big gains with respiratory-muscle, core-stability and high-resistance training. Reviewer's Comment: a balanced view of train-low compete-high on carbohydrate. $http://sportsci.org/2006/wghACSM.htm]Sport and Recreation, AUT University, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?&Hopkins, W G
Schabort, E J
Hawley, J A20012Reliability of power in physical performance tests211-234Sports Medicine31tests, rely?Toussaint, H.M.
Hollander, A.P.1994GEnergetics of competitive swimming - implications for training-programs384-405Sports Medicine186swim, trainreview3journal, plus working from Fig 7 in file under swim? KMeeusen, R.
Duclos, M.
Gleeson, M.
Rietjens, G.
Steinacker, J.
Urhausen, A.2006@Prevention, diagnosis and treatment of the Overtraining Syndrome1-14!European Journal of Sport Science61?
Hopkins, W G2004 An introduction to meta-analysis20-24Sportscience8sCochrane Collaboration, funnel plot, mixed model, quantitative analysis, random effect, research, systematic reviewA meta-analysis is a systematic quantitative review of original research studies of some phenomenon, such as the effect of a specific treatment on some aspect of health or behavior. The meta-analyst expresses the magnitudes of effects from all relevant studies in the same units, then uses an appropriate weighting factor (the inverse of each effect's error variance) to combine the magnitudes into a mean value and its uncertainty (confidence limits). In a traditional meta-analysis, the true effects are assumed to be homogeneous (have the same value) in the analyzed studies, and some "outlier" studies may be eliminated to satisfy this assumption. In the more recent and realistic random-effect or mixed-model meta-analysis, true values of all effects are assumed to be heterogeneous (different), and the analysis provides an estimate of the heterogeneity as a standard deviation representing unexplained typical true variation in the effect between studies. Inclusion of study and mean subject characteristics in the analysis as covariates may reduce heterogeneity and provide further useful information about the magnitude of the effect in different locations and with different subjects. Published effects are usually larger than their true values, owing to the misuse of statistical significance as a criterion for publication. A funnel plot can detect such publication bias, but there is currently no satisfactory way to adjust for it in the meta-analysis, and the only long-term solution is to ban statistical significance.'http://sportsci.org/jour/04/wghmeta.htmSport and Recreation, Faculty of Health, Auckland University of Technology, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz ?Paton, C D
Hopkins, W G2004VEffects of high-intensity training on performance and physiology of endurance athletes25-40Sportscience8Haerobic, anaerobic threshold, economy, plyometrics, resistance, strengthcMost endurance athletes use high-intensity training to prepare for competitions. In this review we consider the effects of high-intensity interval and resistance training on endurance performance and related physiological measures of competitive endurance athletes. METHODS. There were 22 relevant training studies. We classified training as intervals (supramaximal, maximal, submaximal) and resistance (including explosive, plyometrics, and weights). We converted all effects on performance into percent changes in mean power and included effects on physiological measures that impact endurance performance. FINDINGS. All but one study was performed in non-competitive phases of the athletes’ programs, when there was otherwise no high-intensity training. Endurance performance of the shortest durations was enhanced most by supramaximal intervals (~4%) and explosive sport-specific resistance training (4-8%). Endurance performance of the longest durations was enhanced most by intervals of maximal and supramaximal intensities (~6%), but resistance training had smaller effects (~2%). Interval training achieved its effects through improvements of maximum oxygen consumption, anaerobic threshold, and economy, whereas resistance training had benefits mainly on economy. Effects of some forms of high-intensity training on performance or physiology were unclear. CONCLUSIONS. Addition of explosive resistance and high-intensity interval training to a generally low-intensity training program will produce substantial gains in performance. More research is needed to clarify the effects of the various forms of high-intensity training on endurance performance, to determine whether prescribing specific forms of resistance training can improve specific deficits of an endurance athlete's physiology, and to determine the effects of combining the various forms in periodized programs.#http://sportsci.org/jour/04/cdp.htmRCentre for Sport and Exercise Science, The Waikato Institute of Technology, Hamilton. Email: Carl.Paton=AT=wintec.ac.nz. Sport and Recreation, Auckland University of Technology, Auckland 1020, New Zealand. Sport and Recreation, Faculty of Health, Auckland University of Technology, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?Paton, C D
Hopkins, W G2005]Combining explosive and high-resistance training improves performance in competitive cyclists826-830-Journal of Strength and Conditioning Research19?
Batterham, A M
Hopkins, W G2005%A decision tree for controlled trials33-39Sportscience9%analysis, bias, crossover, randomizedFA controlled trial is used to estimate the effect of an intervention. We present here a decision tree for choosing the most appropriate of five kinds of con-trolled trial for numeric outcome measures. A time series or quasi-experimental design is used when there is no opportunity for a separate control group or control treatment. In this design, the weakest of the five, a series of measurements taken before the intervention serves as a baseline to estimate change resulting from the intervention. In trials with a separate control group, the usual design is a fully controlled parallel-groups trial, in which subjects are measured before and after their allocated control or experimental treatment. A posts-only design, in which subjects are measured only after their treatment, can be more efficient when poor reliability of the outcome measure over the time frame of the intervention makes large sample sizes unavoidable. Cross-over studies, in which all the subjects receive all the treatments, are an option when the effects of the treatments wash out in an acceptable time. In fully con-trolled crossovers, subjects are measured before and after each treatment, whereas measurements are taken only after each treatment in a simple cross-over. Fully controlled crossovers, arguably the best of the five designs, are more efficient if the outcome measure becomes too unreliable over the wash-out period, and they provide an assessment of the effect of the treatment on each subject. In simple crossovers, individual assessment is possible only by including a repeat of the control treatment.&http://sportsci.org/jour/05/wghamb.htmSchool of Health and Social Care, University of Teesside, Middlesbrough, UK; Sport and Recreation, AUT University, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?Batterham, A M
Hopkins, W G2005-Making meaningful inferences about magnitudes6-13Sportscience9Bclinical significance, confidence limits, statistical significance8A study of a sample provides only an estimate of the true (population) value of an outcome statistic. A report of the study therefore usually includes an infer-ence about the true value. Traditionally, a researcher makes an inference by declaring the value of the statistic statistically significant or non-significant on the basis of a p value derived from a null hypothesis test. This approach is confusing and can be misleading, depending on the magnitude of the statistic, error of measurement, and sample size. We use a more intuitive and practical approach based directly on uncertainty in the true value of the statistic. First we express the uncertainty as confidence limits, which define the likely range of the true value. We then deal with the real-world relevance of this uncertainty by taking into account values of the statistic that are substantial in some posi-tive and negative sense, such as beneficial and harmful. If the likely range overlaps substantially positive and negative values, we infer that the outcome is unclear; otherwise, we infer that the true value has the magnitude of the observed value: substantially positive, trivial, or substantially negative. We refine this crude inference by stating qualitatively the likelihood that the true value will have the observed magnitude (e.g., very likely beneficial). Quantita-tive or qualitative probabilities that the true value has the other two magnitudes or more finely graded magnitudes (such as trivial, small, moderate, and large) can also be estimated to guide a decision about the utility of the outcome.&http://sportsci.org/jour/05/ambwgh.htmSchool of Health and Social Care, University of Teesside, Middlesbrough, UK; Sport and Recreation, AUT University, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?Batterham, A M
Hopkins, W G2006-Making meaningful inferences about magnitudes50-57:International Journal of Sports Physiology and Performance1M\?Hopkins, W G2007mA spreadsheet for deriving a confidence interval, mechanistic inference and clinical inference from a p value16-20Sportscience11jclinical decision, confidence limits, null-hypothesis test, practical importance, statistical significance#http://sportsci.org/2007/wghinf.htmZSport and Recreation, AUT University, Auckland 0627, New Zealand. Em?
Atkinson, G A2007RWhat’s behind the numbers? Important decisions in judging practical significance12-15Sportscience11cconfidence intervals, null hypothesis, Type I and II statistical errors, smallest worthwhile effect3In an applied field like sport and exercise science, inferences based on estimation of true effect sizes are usually more important than inferences about statistical significance. Inferences about estimation are conventionally made using confidence intervals, which are associated with several critical judgments. The most important decision concerns the smallest effect size that is practically or clinically important. A recently published new approach to sample size estimation also raises issues of judging the appropriate coverage probability of a confidence interval (e.g. 90 or 95%) as well as the degree of overlap between confidence limits and the smallest worthwhile effect. It is these a priori rationalized decisions that underpin the mathematics of confidence intervals, the probabilistic inferences made from them and associated issues like sample size estimation and claims that a statistical approach is too conservative or liberal. First, I discuss that the “null” in the null hypothesis testing process does not always need to be set at zero. If the smallest worthwhile effect itself is selected as the null value, then this process not so isolated from practical significance. Second, I contrast ideas on boundaries of overlap between confidence limits and the smallest worthwhile effect with other published guidelines on using confidence intervals to interpret study results. It is these differences in delimited probability coverage that govern the apparently lower sample sizes required for the new approach. Third, I illustrate how critical the decision on smallest worthwhile effect size can be for accuracy of study conclusions, and question whether uncertainty in this decision process might, in some instances, compromise the accuracy of the inferential statements that are made following statistical analysis. http://sportsci.org/2007/ga.htmResearch Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool L3 2ET, UK. Email: G.Atkinson@livjm.ac.ukd?Hopkins, W G2006@A spreadsheet for combining outcomes from several subject groups51-53Sportscience10>confidence limits, confounding, covariate, inference, modelingData analysis that fails to account for independent groups defined by a subject characteristic (e.g., sex) or by a design characteristic (e.g., treatment order) can result in bias, confounding, and loss of precision in the outcome. Combining the outcomes from separate analyses of the groups is a robust approach to the problem that is easily achieved with the spreadsheet presented here. Differences in the outcome between groups represent the effect of the characteristic on the outcome, while the mean of the outcomes represents the outcome adjusted appropriately for the characteristic. The spreadsheet calculates confidence limits for the differences and for the mean from the confidence limits for the outcome in each group. It also presents magnitude-based inferences for the differences and mean. There are separate cells in the spreadsheet for outcomes represented by means or other normally distributed statistics, relative rates (risk, odds and hazard ratios) or other log-normally distributed statistics, and correlation coefficients.#http://sportsci.org/2006/wghcom.htm]Sport and Recreation, AUT University, Auckland 0627, New Zealand. Email: will=AT=clear.net.nz?Fisher, R A1921QOn the probable error of a coefficient of correlation deduced from a small sample3-32Metron1
stats ???O?Hopkins, W G2006\Spreadsheets for analysis of controlled trials, with adjustment for a subject characteristic46-50Sportscience10fcrossover, design, inference, repeated measures, intervention, randomized, transformation, t statisticSpreadsheets previously available at this site for analysis of controlled trials have been updated to allow inclusion of one covariate representing a subject characteristic. The spreadsheets provide estimates of the effect of an intervention adjusted to any chosen value of the covariate, thereby reducing the possibility for confounding of the effect when a characteristic such as age, fitness or sex is unequal in the experimental and control groups. The pre-test value of the dependent variable can also be included as a covariate to avoid confounding by the phenomenon of regression to the mean. Graphs of change scores plotted against the covariate show visually how the treatment effect is adjusted to the chosen value of the covariate. The spreadsheets also provide an estimate of the effect of the covariate itself, representing individual responses attributable to the covariate. Other new features of the spreadsheets include plots of raw and back-transformed means with easily modified standard-deviation bars, and qualitative inferential outcomes based on interpretation of the span of the confidence interval relative to magnitude thresholds for trivial, small, moderate, large, and very large.(http://sportsci.org/2006/wghcontrial.htm]Sport and Recreation, AUT University, Auckland 0627, New Zealand. Email: will=AT=clear.net.nzv?Perneger, T. V.1998(What's wrong with Bonferroni adjustments 1236-1238BMJ316Br Med Assoc ail: will@clear.net.nz?Hopkins, W G20048How to interpret changes in an athletic performance test1-7Sportscience8Bayes, correlation, error of the estimate, error of measurement, limits of agreement, reliability, time to exhaustion, time trial, validitybWhen monitoring progression of an athlete with performance or other fitness tests, it is important to take into account the magnitude of the smallest worth-while change in performance and the uncertainty or noise in the test result. For elite athletes competing in sports as individuals, the smallest worthwhile change in performance is about half the typical variation in an athlete's per-formance from competition to competition, or ~0.5-1% when expressed as a change in power output, depending on the sport. In team sports, where there is no direct relationship between team and test performance, an appropriate default for the smallest change in test performance is one-fifth of the between-athlete standard deviation (a standardized or Cohen effect size of 0.20). Noise in a test result is best expressed as the typical or standard error of measurement derived from a reliability study. The noise in most perform-ance tests is greater than the smallest worthwhile difference, so assessments of changes in performance can be problematic. An exact but somewhat impractical solution is to present chances that the true change is beneficial, trivial, and harmful. A simpler approach is to apply systematic rules to decide whether the true change is beneficial, trivial, harmful, or unclear. Unrealistically large changes can also be partially discounted when tests are noisy.(http://sportsci.org/jour/04/wghtests.htmSport and Recreation, Faculty of Health, Auckland University of Technology, Auckland 1020, New Zealand. Email: will=AT=clear.net.nz?Hopkins, W G2007,A spreadsheet to compare means in two groups22-23Sportscience11)http://sportsci.org/2007/inbrief.htm#xcl2ZSport and Recreation, AUT University, Auckland 0627, New Zealand. Email: will@clear.net.nz?1Hopkins, W G
Marshall, S W
Quarrie, K L
Hume, P A20074Risk factors and risk statistics for sports injuries208-210"Clinical Journal of Sport Medicine17O?Cohen, J19886Statistical power analysis for the behavioral sciences
Hillsdale, NJLawrence Erlbaum2nd!stats, design, effect size, powerpartPKQ
7I/**refs.FRM 0B< !// !HPRIMARYyearIndex
6ByP/) idreference_typetext_stylesauthoryear title
pagessecondary_titlevolume
numbernumber_of_volumessecondary_authorplace_published
publishersubsidiary_authoredition keywords
type_of_workdate2) abstractlabelurltertiary_titletertiary_author notes
isbn custom_1 custom_2
custom_3 custom_4alternate_titleaccession_numbercall_numbershort_title custom_5 custom_6sectionoriginal_publicationH) reprint_editionreviewed_itemauthor_addressimagecaption custom_7
electronic_resource_numberlink_to_pdftranslated_author
translated_titlename_of_databasedatabase_providerresearch_notes languageaccess_datelast_modified_date
!!H!H!H! (H!
3H!>H!IH!
TH!_H!jH!uH!
H!H!H! H!
H!H! H!H!H!H!H! H!
H! H! H!
%H! 0H!;H!FH!QH!\H! gH! rH!}H!H!H!H!H!H!H! H!
H!H!H!
H!H!H!H! "H!-H!8H!idreference_typetext_stylesauthoryeartitlepagessecondary_titlevolumenumbernumber_of_volumessecondary_authorplace_publishedpublishersubsidiary_authoreditionkeywordstype_of_workdateabstractlabelurltertiary_titletertiary_authornotesisbncustom_1custom_2custom_3custom_4alternate_titleaccession_numbercall_numbershort_titlecustom_5custom_6sectionoriginal_publicationreprint_editionreviewed_itemauthor_addressimagecaptioncustom_7electronic_resource_numberlink_to_pdftranslated_authortranslated_titlename_of_databasedatabase_providerresearch_noteslanguageaccess_datelast_modified_datePK}j7z4VVrefs.MYDPKQ
7I/**Wrefs.FRMPKl