Design & Analysis of Research in HPERD
Design & Analysis of Research in HPERD PEP 455
Popular in Course
Popular in Physical Education
This 9 page Class Notes was uploaded by Brooklyn Haag on Friday October 23, 2015. The Class Notes belongs to PEP 455 at University of Idaho taught by Staff in Fall. Since its upload, it has received 29 views. For similar materials see /class/227791/pep-455-university-of-idaho in Physical Education at University of Idaho.
Reviews for Design & Analysis of Research in HPERD
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/23/15
Chapter Outline Probability Meaningfulness Power Using information in the context of the stud Reporting statistical data Truth Table for Interpreting Statistical Findings the Null Hypothesis Pro mun level my occunume hype I umrl Ha We Ha fase pical m p lt e Varying alph Accept Currectdemsmn Type M Errur Truth table tbefa Exact thahi39w Reject Type l Errur Curran demem eaem mm mm Meaningfulness Power proha of rejecting the hypothesis when s false Sampling for Nu Hypothesis context of the Study mm 1 an MT 5 39 How do ndings from the study t within the context of Practice cm rm summymmwmmmmymm 3mm Iawa mmawvs Revnmedwnh Megawath an mpmmwaswmcm ammmmmmm m xnnnnz l5 2 Planning Research Information needed in planning Effect size Power Reporting Statistical Data Summary From APA and APS Always report screen your data Select minimally statistical analyses Report Report magnitudes of the effects Report variability using Report data to appropriate level Sample Size Regardless of size the crucial factor is whether or not the sample is of the population thus how the sample is selected Points to consider regarding sample size Nature of the study Statistical considerations Variability ofpopulation Number oftreatment groups Practical factors Nature of the study Descriptive Correlational or Difference Descriptive and Correlational studies typically should have research participants Difference studies o en employ research participants Statistical considerations How do you want to analyze the data What statistical application will be used Power of the statistical test Power is the probability that the test will the Hnwhen in fact the HEl is false In general the the sample size the more power of the statistic being used Generally N30 is minimum needed to meet assumptions of many statistical procedures Variability of population Sample size is related to sampling error The the sample size the the samp 9 error and the greater likelihood sample is ofthe population variability small sample will suffice High variability sample size will be Number of Treatment Groups When samples are divided into smaller groups to be compared it is important that the subgroups are of Should be more concerned with cell size than total sample size Practical Factors Availability of research participants Costs Time Comments About Sampling Descriptive and correlational research are vitally concerned about the representativeness of the sample usually necessitating larger sample sizes and more attention iven o e sam in I rocedure Difference studies can often get by with sample sizes as long as internal is maintained In practice volunteer research participants are involved in a ood portion of research Be aware of the potential of systematic error being introduced in the study Random Assignment The purpose is to establish grou before the introduction of the independent variable Two basic methods groups design Repeated measures design Independent Groups Design Each research participant is randomly assigned to of the various treatment groups Each subject participates in group Repeated Measures Design Subjects participate in more than one group treatment condition In the simplest example each research participant would be assigned to each level of the independent variable and then is measured after receiving the treatment Counterbaancng IS often used 0 conio folpoSSbe order effect Summary Planning a Study Issues to attend to Statistical Signi cance Type 1 and 2 Errors Power Sample size As it reiates to type of study Independent and Dependent study designs Correlation A statistical technique that is used to measure and describe a between Variables are observed not controlled or manipulated Requires scores for Often uses a scatterplot individual X amp Y Attributes of r Nmnlr I a 395 2 me Zarc Low Correlation and the Are two variables related other one changes Examples Skinfolds and percent body fat Time to run 15 miles and VOzmax Pearson ProductMoment What happens to one variable when the Amount of study time and performance on a test Tire tread and amount of time the car radio is on Speed ofa car and the likelihood of getting a ticket Scatte rp l ot n D u A plot used to provide a visual observation of the between two variables Vvolues K unlues Scatterplot of Correlation Pull Ups and ChinUps D E 15 39 E a g A I g 1039 o O E o m g C n 7 I E 5 U39 4 1D Pull ups nur n39ber compieced 3915 Scatterplot of Correlation Body Weight and Pull Ups 1 5 a E E 1 D e 8 quot5 o E o E quotS 5 e 39 39 a D I n 1 I x 120 1 30 1140 1 SO 1130 170 180 Weight Ijlb Scatterplot of Zero Correlation r O Eb Yne o g 2 o l l l 39D a 4 5 El 19 x Correlation PPM Formula XXVquot11 XXIIIZV erDEEXZ 2523 HZV2 25321 Correlation Measures three characteristics of the relationship between the variables of the relationship of the relationship of the relationship Correlation Direction of the relationship Identified by the coefficient r Positive the variables tend to move in the same direction Negative the variables tend to move in opposite directions of the correlation Correlation Positive Correlation Negative Correlation 4 r r 9 39 391 I quot 5quot 39 39 739 quot w ii 3 e 2 2G 3 V quot r r Z 3 v L o 7 J 7 a 39 5 7 Correlation Correlation Degree of the relationship How well the data points fit the form being considered T Straight Line ypes Pearson Product Moment Correlation Moderate No LOW Perfect High 7 Used to measure the between 3 quot two variables l 39 V Used with ratio or interval data m 39 L 39 i m Rank Ordered Correlation 7 quot a Q I quot39 Used to measure the linear degree between two 39 m sets of ranks 39 39 39 Used with data rs A 39r 77 39 er T39r Vf quot zquot Factors to Consider in the Factors to Consider in the Interpretation of Correlation Interpretation of Correlation Affected by the of the scores Describes the relationship only Does not suggest a cause effect relationship Factors to Consider in the Interpretation of Correlation l imitations of r 5 OutridersOutliers E one or two extreme data points can dramatically quotS 4 influence the value of the correlation coefficient E E 3 2 18 E 2 J 39 E 1 A 7 39 l I l l quotIquot 217 2 o 2 4 as 8 1o 7 39 H Anxiety arbitrary units E Coefficient of determination ttr 2 Represents the amount of that the two variables Or that the variance within the two variables taps a element Eg If we correlated arm strength and of chin ups and found r r2 or 640A shared variance Or that 360A remains unexplained Suggesting that accounts for 0A of the ability to do chin ups Differences between ttr amp ttr 2 r ives us the correlation between 2 variables r2 ives us the blt 2 variables r2 allows us to make based on An r2 of 50 is as bi as an r2 of 25 the same is not true for r since r2 is a measure of shared variance Factors to Consider in the Interpretation of Correlation o Coefficient of Determination r2 measures the proportion of variability in one variable that can be explained determined from the relationship with the other variable De enaem variable Independent variable p predided Dredlctor Regression 0 Statistical technique A 60 tx for fIndIng the g 1 a 50 z u n 5 no U stralght llne for a set g 4039 of data E 30 1 5 1 o The llne that IS found E 20 B I IS called the g 10 ne 43920 30 40 so so 70 5E7 Te mosrclure degrees F Regression o Purposes of the regression line Make the easier to Identifies the or of the relationship Establishes a precise relationship that can be used for regresson 0 Each regression line can be expressed by an equation Regression Line of Best Fit Y bX a a and b are fixed constants b slope change in Y as a result of a change in X b r Sy SX a y intercept value of Y when X O a bx Y estimated value Correla tion and Prediction Skit Ifolds Correla tion and Prediction 77 Skinfolcls 4quot enoe intemals from Chapter 3 we would 1n fold of 40 would equal body l EI SEE 3e SPSS Correla tion Example Correlations r Correlatioh ie at the 001 level 2etailed