• Home
  • Swinburne University of Technology
  • Foundations of statistics
  • Exploring Relationships and Statistical Tests

Exploring Relationships and Statistical Tests

Week 11 Lecture SWIN BUR NE SWINBURNE UNIVERSITY OF TECHNOLOGY STA10003 FOUNDATIONS OF STATISTICS Sandra Wyatt KNOW ING Revision of Week 10 · Exploring relationships between two metric variables · Scatterplots ... Check for linearity, sub-groups, outliers · Pearson's correlation co-efficient [Pearson's r ] . Used to measure the strength of the linear relationship · 95% Confidence Interval for Pearson's r [strength in population] . Co-efficient of determination [r 2] . Gives an indication of the 'importance' of the IV in predicting the DV · Regression Equation · Y = bX + a . Used to predict the value of the DV if we know something about the IV · Report writing! 2 1 Revision of Week 10 0 0 Assignment Mark [out of 20) 0 0 0004 0 0 Number of Tutorials Attended (up to / including Week 7] r2 = 0.112 Correlations Assignment Mark [out of 20] Number of Tutorials Attended Pearson Correlation 1 .335 ** Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) Number of Tutorials Attended [up to / including Assignment Mark [out of 20] Week 7] .000 312 312 .335 ** 1 .000 [up to / including Week 7] N 312 312 **. Correlation is significant at the 0.01 level (2-tailed). 95% Confidence Interval: (0.23, 0.43) Model 1 (Constant) Coefficientsª Standardiz ed Coefficient s Unstandardized Coefficients Std. Error Beta B 8.884 .661 t 13.439 Number of Tutorials Attended [up to / including Week 7] .723 .116 .335 6.253 .000 95.0% Confidence Interval for B Lower Bound 7.583 Upper Bound 10.185 Sig. .000 .496 .951 a. Dependent Variable: Assignment Mark [out of 20] Note: This information relates to a previous cohort of STA10003 students IING 3 Report Writing It was hypothesised that STA10003 statistics students who attend more tutorials tend to achieve higher marks for the Assignment. In this cohort of STA10003 students [who submitted an Assignment], there was a moderate strength, positive, linear relationship between Assignment marks achieved and the number of tutorials attended, and Pearson's r shows that this relationship is significant (r = . 34, n = 312, p < . 001). The 95% confidence interval for Pearson's r indicates that the strength of the relationship is between p = . 23 and p = . 43. For each additional tutorial attended, on average, Assignment marks were 0.72 higher / more. Conclusion? 0 r2 interpretation: 11.2% of the variation is Assignment marks can be explained by the linear relationship between tutorial attendance and Assignment marks. 4 2 Overview . Parametric and Non-Parametric Tests · Crosstabulations and Chi-Square [12] · Chi-Square Test for Goodness of Fit • For information purposes / background · Chi-Square Test for Independence · Report writing · Putting together our crosstabulations and Chi-Square information - What to report ... · Risk / Odd Ratios KNOW ING 5 Parametric and Non-Parametric Tests · Parametric Tests · Numerical scores · Assumptions about population parameters · eg population means / mean difference · normality of population distributions · Non-Parametric Tests · Categories or groups