logo

Correlation and ANOVA Analysis of Child Data

   

Added on  2020-04-15

10 Pages1611 Words112 Views
 | 
 | 
 | 
PSYC2012 - PSYCHOLOGY: AN EVIDENCE-BASED APPROACH 2ASSESSMENT 2 Data Analysis AssignmentQuestion oneParametric test assumes that the data is normally distributed, which means that the peak is at the centre and symmetric about the measure of location that is mean or median. Normally assumptions eliminate outlier and extreme values which give false inferences about the distribution of data. Test for normality includes histogram, Q-Q plot and comparing kurtosis and skewness.Test of normality using Shapiro-Wilk test for normality at 5% level of significanceNull hypothesis: The data is normally distributed.Alternative hypothesis: The data is not normally distributed.Tests of NormalityKolmogorov-SmirnovaShapiro-WilkStatisticdfSig.StatisticdfSig.the results of a simple reaction time test for each child.1427.00.200*.9727.00.65the results of a qualitative EEG brain scan of each child’s brainwave patterns.2227.00.00.8827.00.00child’s results on the Weschler Child Memory Scale.1427.00.200*.9127.00.02parental ratings of levels of difficult behaviour exhibited .1427.00.16.9727.00.60
Correlation and ANOVA Analysis of Child Data_1

by each childresults of modified Stanford-Binet IQ testing of each child.1327.00.200*.9627.00.29child’s results for asimple learning test.1227.00.200*.9627.00.28child’s results on atest of verbal ability.1127.00.200*.9827.00.86results of an experimental adaptive behaviour scale developed by the researcher.1327.00.200*.9627.00.45*. This is a lower bound of the true significance.a. Lilliefors Significance CorrectionThe p-value for the results of a qualitative EEG brain scan of each child’s brainwave patterns is 0.00 which indicates strong evidence of non-normality. All other variable have p-value greater than 0.05 (level of significance) thus normally distributed. The variable can be improved by transformation which includes:a)Log transformation, finding the logarithm of variableTests of NormalityKolmogorov-SmirnovaShapiro-WilkStatisticdfSig.StatisticdfSig.logx2.1427.20.9527.26a. Lilliefors Significance CorrectionThe data now is normally distributedb)Finding the square root of the variable
Correlation and ANOVA Analysis of Child Data_2

Tests of NormalityKolmogorov-SmirnovaShapiro-WilkStatisticDfSig.StatisticdfSig.squarex2.2627.00.7527.00a. Lilliefors Significance CorrectionThis did not improve the data after squaring the data is still non-normal.c)Finding the square of variable Tests of NormalityKolmogorov-SmirnovaShapiro-WilkStatisticdfSig.StatisticdfSig.sqrtx2.1927.02.9427.10a. Lilliefors Significance CorrectionThe transformations improve the data and now are normal.Question ThreeChi-square assumes the following assumptions:The data must be categorical in nature that is nominal or ordinalThe variable must have more than two independents categoriesChi-Square TestsValuedfAsymp. Sig.(2-sided)Pearson Chi-Square7.00a2.03Likelihood Ratio9.642.01Linear-by-LinearAssociation6.021.01N of Valid Cases27
Correlation and ANOVA Analysis of Child Data_3

End of preview

Want to access all the pages? Upload your documents or become a member.

Related Documents