2Definitions and Mis/interpretations Planning Performing Publishing OutlineDefinitions and Mis/interpretationsPlanningSample sizePerformingSample size "on the fly"PublishingMethods, Results, DiscussionMeta-analysisPublishing non-significant outcomesConclusionsDis/advantages
3Definitions and Mis/interpretations Confidence limits: Definitions"Margin of error"Example: Survey of 1000 voters Democrats 43%, Republicans 33% Margin of error is ± 3% (for a result of 50%...)Likely range of true value"Likely" is usually 95%."True value" = population value = value if you studied the entire population.Example: Survey of 1000 voters Democrats 43% (likely range 40 to 46%) Democrats - Republicans 10% (likely range 5 to 15%)
4correlation coefficient Example: in a study of 64 subjects, the correlation between height and weight was 0.68 (likely range 0.52 to 0.79).observedvaluelowerconfidencelimitupperconfidencelimit0.000.501correlation coefficient
5Confidence interval: difference between the upper and lower confidence limits. Amazing facts about confidence intervals (for normally distributed statistics)To halve the interval, you have to quadruple sample size.A 99% interval is 1.3 times wider than a 95% interval. You need 1.7 times the sample size for the same width.A 90% interval is 0.8 of the width of a 95% interval. You need 0.7 times the sample size for the same width.
6How to Derive Confidence Limits Find a function(true value, observed value, data) with a known probability distribution.Calculate a critical value, such that for 2.5% of the time, function(true value, observed value, data) < critical value.probabilityfunction (e.g. (n-1)s2/2)area = 0.025critical valueprobability distribution of function (e.g. 2)Rearranging, for 2.5% of the time, true value > function'(observed value, data, critical value) = upper confidence limit
7Mis/interpretation of confidence limits Hard to misinterpret confidence limits for simple proportions and correlation coefficients.Easier to misinterpret changes in means.Example: The change in blood volume in a study was 0.52 L (likely range 0.12 to 0.92 L).For 95% of subjects, the change was/would be between 0.12 and 0.92 L.The average change in the population would be between 0.12 and 0.92 L.The change for the average subject would be between 0.12 and 0.92 L.There may be individual differences in the change.
8correlation coefficient P value: DefinitionThe probability of a more extreme absolute value than the observed value if the true value was zero or null.Example: 20 subjects, correlation = 0.25, p = 0.29.probabilitycorrelation coefficientarea = p value= 0.29no effectobserved effect(r = 0.25)distribution ofcorrelationsfor no effectand n = 200.5-0.5
9correlation coefficient "Statistically Significant": DefinitionsP < 0.05Zero lies outside the confidence interval.Examples: four correlations for samples of size 20.0.000.501correlation coefficient-0.50rlikely rangeP0.700.37 to 0.870.0070.440.00 to 0.740.050.25-0.22 to 0.620.290.00-0.44 to 0.441.00
10Incredibly interesting information about statistical significance and confidence intervals p < 0.05p = 0.05p > 0.05Two independent estimates of a normally distributed statistic with equal confidence intervals are significantly different at the 5% level if the overlap of their intervals is less than 0.29 (1 - 2/2) of the length of the interval.If the intervals are very unequal...p < 0.05p = 0.05p > 0.05
11Type I and II ErrorsYou could be wrong about significance or lack of it.Type I error = false alarm.Rate = 5% for zero real effect.Type II error = failed alarm.Traditional acceptable rate = 20% for smallest worthwhile effect.Lots of tests for significance implies more chance of at least one false alarm: "inflated type I error".Ditto type II error?Deal with inflated type I error by reducing the p value.Should we adjust confidence intervals? No.
12Mis/interpretation of P < 0.05 (for an observed positive effect) The effect is probably big.There's a < 5% chance the effect is zero.There's a < 2.5% chance the effect is < zero.There's a high chance the effect is > zero.The effect is publishable.Mis/interpretation of P > 0.05 (for an observed positive effect)The effect is not publishable.There is no effect.The effect is probably zero or trivial.There's a reasonable chance the effect is < zero.
13Sample Size via Statistical Significance Planning ResearchSample Size via Statistical SignificanceSample size must be big enough to be sure you will detect the smallest worthwhile effect.To be sure: 80% of the time.Detect: P < 0.05.Smallest worthwhile effect: what impacts your subjectscorrelation = 0.10relative risk = 1.2 (or frequency difference = 10%)difference in means = 0.2 of a between-subject standard deviationchange in means = 0.5 of a within-subject standard deviationExample: 760 subjects to detect a correlation of 0.10.Example: 68 subjects to detect a 0.5% change in a crossover study when the within-subject variation is 1%.
14But 95% likely range doesn't work properly with traditional sample-size estimation (maybe). Example: Correlation of 0.06, sample size of47.5% % (=95%) likely range:0.1correlation coefficient-0.1Not significant, but could be substantial.Huh?0.1correlation coefficient-0.147.5% + 30% likely range:Not significant, and can't be substantial.OK!
15correlation coefficient Sample Size via Confidence LimitsSample size must be big enough for acceptable precision of the effect.Precision means 95% confidence limits.Acceptable means any value of the effect within these limits will not impact your subjects.Example: need 380 subjects to delimit a correlation of zero.0.10correlation coefficient-0.10smallest worthwhileeffectsconfidenceinterval forN = 380
16correlation coefficient But sample size needed to detect or delimit smallest effect is overkill for larger effects.Example: confidence limits for correlations of 0.10 and 0.80 with a sample size of0.10.30.50.70.91correlation coefficient-0.1So why not start with a smaller sample and do more subjects only if necessary? Yes, I call it...
17correlation coefficient Performing ResearchSample Size "On the Fly"Start with a small sample; add subjects until you get acceptable precision for the effect.Acceptable precision defined as before.Need qualitative scale for magnitudes of effects.Example: sample sizes to delimit correlations...15126.96.36.199.70.91trivialsmallmoderatelarge270350380correlation coefficient-0.1nearlyperfect46very large
18Problems with sampling on the fly Do not sample until you get statistical significance: the resulting outcomes are biased larger than life.Sampling until the confidence interval is acceptable produces bias, but it is negligible.But researchers will rush into print as soon as they get statistical significance.And funding agencies prefer to give money once (but you could give some back!).And all the big effects have been researched anyway? No, not really.
19In the Methods Publishing Research "We show the precision of our estimates of outcome statistics as 95% confidence limits (which define the likely range of the true value in the population from which we drew our sample)."Amazingly useful tips on calculating confidence limitsSimple differences between means: stats program.Other normally distributed statistics: mean and p value.Relative risks: stats program.Correlations: Fisher's z transform.Standard deviations and other root mean square variations: chi-squared distribution.
20coefficient of variation (%) Coefficients of variation: standard deviation of 100x natural log of the variable. Back transform for CV>5%.Use the adjustment of Tate and Klett to get shorter intervals for SDs and CVs from small samples.21coefficient of variation (%)3Example: coefficient of variation for 10 subjects in 2 testsusualadjustedRatios of independent standard deviations: F distribution.R2 (variance explained): convert to a correlation.Use the spreadsheet at sportsci.org/stats for all the above.Effect-size (mean/standard deviation): non-central F distribution or bootstrapping.Really awful statistics: bootstrapping.
21Bootstrapping (Resampling) for confidence limits Use for difficult statistics, e.g. for grossly non-normal repeated measures with missing values. Here's how...For a large-enough sample, you can recreate (sort of) the population by duplicating the sample endlessly.Draw 1000 samples (of same size as your original) from this population.Calculate your outcome statistic for each of these samples, rank them, then find the 25th and 975th place-getters. These are the confidence limits.ProblemsPainful to generate.No good for infrequent levels of nominal variables.
22In the ResultsIn TEXTChange or difference in means First mention: (95% confidence/likely limits/range to 0.93) or (95% confidence/likely limits/range ± 0.51). Thereafter: (1.4 to 3.8) or 2.6 (± 1.2) etc.Correlations, relative risks, odds ratios, standard deviations, ratios of standard deviations: can't use ± because the confidence interval is skewed: ...a correlation of 0.90 (0.67 to 0.97) a coefficient of variation of 1.3% (0.9 to 1.9)...
23In TABLES Confidence intervals r likely range 0.70 0.37 to 0.87 0.44 0.25-0.22 to 0.620.00-0.44 to 0.44Variable AVariable BVariable CVariable DP valuesrp0.700.0070.440.050.250.290.001.00Variable AVariable BVariable CVariable DAsterisksr0.70**0.44*0.250.00Variable AVariable BVariable CVariable D
24In FIGURES-10-5510Change in power (%)Bars are 95% likely rangesTold placeboNot toldTold carbohydrate
254 sea level altitude sea level 3 live low train low 2 1 change in 5000-mtime (%)live high train highlikely range oftrue change-1-2live high train low-32468101214training time (weeks)
26correlation coefficient In the DiscussionInterpret the observed effect and its 95% confidence limits qualitatively.Example: you observed a moderate correlation, but the true value of the correlation could be anything between trivial and very strong.0.10.30.50.70.91trivialsmallmoderatelargecorrelation coefficient-0.1nearlyperfectvery large
27Meta-AnalysisDeriving a single estimate and confidence interval for an effect from several studies.Here's how it works for two:Study 1Study 2Study 1+2Equal Confidence IntervalsStudy 1Study 2Study 1+2Unequal Confidence Intervals
28Publishing non-significant outcomes Publishing only significant effects from small-scale studies leads to publication bias.Publishing effects with confidence limits regardless of magnitude is free of bias.Many smaller studies are probably better than a few larger ones anyway.So bully the editor into accepting the paper about your seemingly inconclusive small-scale study.
29Disadvantages of Statistical Significance ConclusionsDisadvantages of Statistical SignificanceEmphasizes testing of hypotheses.Aim is to detect an effect--effects are zero until proven otherwise.Have to understand Type I and II errors.Hard to understand; easy to misinterpret.Have to consider sample size.Focuses on statistically significant effects.Advantages of Statistical SignificanceFamiliar.All stats programs give p values.Easy to put asterisks in tables and figures.
30Disadvantages of Confidence Limits Unfamiliar.Not always available in stats programs.Cluttersome in tables.Display in time series can be a challenge.Advantages of Confidence LimitsEmphasizes precision of estimation.Aim is to delimit an effect--effects are never zero.Only one kind of "error".Meaning is reasonably clear, even to lay readers.No confusion between significance and magnitude.Journals now require them.