Statistical Analysis. Means and standard deviations for body composition measures were calculated for each athlete sub-group. These included total mass, lean mass, and fat mass for the whole body, as well as for the trunk, leg and arm regions. Data from repeated scans were used to calculate change in the mean (the mean difference between the repeated scan results), typical error of the measurements (TEMs; standard deviation of the difference scores of all athletes in the group divided by √2, in grams and %) and intraclass correlation coefficients (ICCs) for all body composition measures, using a published spreadsheet (Hopkins 2000b). To ensure normality of the sampling distribution, each of these measurements were firstly log transformed before analysis and back transformed after analysis, as recommended by Hopkins (2000a). TEMs were derived for the whole cohort, for each sub-group of athletes (each sport separated by gender; n = 7) and for male and females. To test whether the TEMs differed by height, weight or body fat percentage, TEMs were computed for the first and fourth quartiles when athletes were ranked according to each of these descriptors. Uncertainty in the TEM estimates were expressed as 90% confidence limits (CL). The typical error differences between the two groups for each demographic (gender, height, weight and body fat percentage) were considered clear, if the 90% confidence intervals (CI) of the groups did not overlap. Additionally, Pearson correlation coefficients were used to assess the relationship between the mean fat masses and the fat mass TEMs of the associated body regions. According to Hopkins (2000a), the TEM (which represents the error in both directions) should be multiplied by a factor of 1.5 to 2 before interpreting longitudinal changes. Thus, TEMs were doubled to provide a conservative ‘TEM threshold’ above which changes were considered likely (92%probability) to be ‘true’ changes. Data from the first scans were used as an estimate of baseline body composition. For the follow-up DXA scans, percentage changes (from baseline and between time points) in three whole body composition measures (total body mass, lean mass and fat mass) were calculated for all bob skeleton athletes and rugby players at each time point. Additionally, for the bob skeleton athletes only, percentage changes in leg lean mass were calculated at each time point as the emphasis of training was lower limb hypertrophy. The percentage changes in total lean mass, leg lean mass, an...

Statistical Analysis. Means and standard error of the mean were calculated for the mycelial growth inhibition and germinated seeds after composts teas treatment measured for the three sets of experiments in each case. These means were statistically compared using the LSD Fischer test was used to determine if they were significantly different at P< 0.05.Cm : Cattle manureII. RESULTS

Statistical Analysis. Means, and standard deviations for each characteristic will be calculated. Paired sample t-test will be computed to assess changes in before treatment and follow-up scores. Statistical significance will be calculated and two-tail significance level of 0.05 will be used. All analyses will be conducted using IBM SPSS 21.0.

## Examples of *Statistical Analysis* in a sentence

If you have any comments about these estimates or any other aspects of this data collection, contact: US Department of Labor, OSHA Office of

**Statistical Analysis**, Room N-3644, 200 Constitution Avenue, Company executive Title ( ) / / NW, Washington, DC, 20210.Prior to the analysis of the final study data, a detailed

**Statistical Analysis**Plan (SAP) will be written describing all analyses that will be performed.Clarke D.C., M.K. Morris, and D.A. Lauffenburger, “Normalization and

**Statistical Analysis**of Multiplexed Bead- Based Immunoassay Data using Mixed-Effects Modeling”, Molec.Nash,

**Statistical Analysis**of Sample Data from Tank 48H, WSRC-TR-95-0325, Rev.Full detail is in the

**Statistical Analysis**Plan (SAP) (Section 8.2).

## More Definitions of *Statistical Analysis*

Statistical Analysis. Means, standard deviations and confidence intervals were recorded for kick leg linear foot speed and lower body kinematics at ball contact as well as maxima, minima and range of motion for the kick leg and support leg. Repeated measures ANOVAs were performed in SPSS 20 to evaluate the difference between sprint times and the influencefatigue had on each variable. Significance was set at P < .05 and effect sizes (partial ƞ2)were calculated for comparison. Cohen (1992) suggested effect sizes for various indexes, including ƞ2 (small = .0099, medium=.0588, large=.1379). However, when thedegrees-of-freedom of the numerator exceeds 1, as it did with each variable in this study, eta-squared is compared to R-squared (Levine & Hullett, 2002). So adapting Cohen’s (1992) thresholds, equivalent classification for effect sizes were used as the square root of these thresholds (.01 is a small effect, 0.09 a medium effect and 0.25 a large effect) (Pierce, Block & Aguinis, 2004). Post hoc analysis was performed on all significant variables. Maximum errors were calculated for each variable (based on 100Hz vs 500Hz data from a previous study in the lab) and those no longer significant with maximum possible errors were not included for discussion. Least-significant-difference multiple comparison procedure (Rahnama et al., 2003), were used to determine the specific differences between each fatigue cycle data. Correlation coefficients (r) were also calculated to evaluate the relationship between foot speed and each significant dependent variable to determine if the relationship changed under fatigue. Those that were higher than the critical value of 0.707 (df = 6, P = 0.05) were included in the discussion (Bluman, 2004).

Statistical Analysis. Means and standard errors were calculated for each experiment and the data were compared using the ANOVA test and the significance between means was compared by LSD values at 0.05 level, using SAS program (SAS institute, 1988).

Statistical Analysis. Means and standard deviations for the lower limb muscle activity and ankle alignment data across the five walking trials were calculated per boot condition. A three-way repeated measures ANOVA design, with three within factors of boot shaft type (flexible and stiff), soletype (flexible and stiff), and surface condition (gravel and soft) was then used to determine whether there were any significant main effects or interactions of either shaft type, sole type, or surface condition on the lower limb muscle activity and lower limb motion data displayed by the participants. Wilks' Lambda multivariate test was used to determine any significant main effects and interactions. Paired t-tests further investigated any significant main effects and interactions. This design determined whether any of the data were significantly different between the boot shaft and sole types and whether any of these differences were influenced by which surface the participants were walking on. An alpha level of p ≤ 0.05 was used for all statistical comparisons and all tests were conducted using SPSS statistical software (Version 21, SPSS, USA).

Statistical Analysis. Means and standard deviations were calculated for continuous variables. Wilcoxon rank sum tests were used to test for differences between groups (e.g., result of lawsuit, changes in medical practice) in their responses to the scales. Spearman rank correlation coefficients were used to assess the degree of association between questionnaire scales (e.g., cooperation of participants, awareness of changes in medical care). Surveys with missing values were not excluded; all item-level responses were incorporated into the analysis.SPSS Statistics Version 19.0 (SPSS Inc., Chicago, USA) was used for all analyses. Statistical significance was set at a p-value of 0.05.

Statistical Analysis. Means, standard deviations (SD), medians, and interquartile ranges (IQR) were computed to describe the data. A customisable statistical spreadsheet and between-participant pre-race SD were used to compute effect sizes (ES) [16]. The smallest worthwhile difference in means was set to 0.20 of these SDs, except for foot-strike angle where it was set to 2.5˚ based on prior test- retest data [27]. Magnitudes of the ES were interpreted as trivial (ES < 0.2), small (0.2 ≤ ES < 0.6), moderate (0.6 ≤ ES < 1.2), and large (ES ≥ 1.2), and deemed clear if their 90% confidence interval [lower, upper] did not overlap thresholds for small positive and small negative effects. Variables were log-transformed to reduce bias arising from non-uniformity of error and used for interpreting all statistical comparisons, except for foot-strike angle where log- transformation was not appropriate. Statistical significance from paired t-tests was set at P <ACCEPTED MANUSCRIPT0.05. The 3-km and 10-km foot-strike angles were compared using the same statistical approaches. Levels of agreement and 90% confidence intervals between pre-race and post-race, 3-km and 10-km, and perceived and actual foot-strike patterns were computed using the Wilson score method incorporating continuity correction [9].

Statistical Analysis. Means for growth, proximate composition, and fish body fatty acids were analyzed using one-way Anova, after verifying the homogeneity of their variance [24]. Values for percentage data and ratios were log-transformed prior to analyses. When the effect was significant, compari- sons between treatment means was run using Duncan’s multiple range test [25] at P = 0.05. All analysis were done using SPSS program version 17.0 (SPSS, Chicago, Illinois, USA).

Statistical Analysis. Means and standard deviations were determined for each condition studied. Gap marginal data were normally distributed as determined by the Kolmogorov-Smirnov test so one-way analysis of variance (ANOVA) followed by Tukey test was applied, with the significance level significance of 5%.