{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

OneWayANOVA_SP08_BB_nocolor

# OneWayANOVA_SP08_BB_nocolor - Test Guide 2 Groups...

This preview shows pages 1–5. Sign up to view the full content.

1 Test Guide 2 Groups/ Conditions > 2 Groups/ Conditions Do independent groups differ on one mean? Independent/ between subjects t-test One-way ANOVA Does one group differ across repeated measures? Paired/dependent/ within subjects t-test One-way Repeated measures ANOVA Multiple independent variables ~ main effects and interactions Factorial ANOVA Factorial ANOVA Test Guide One Way… – One factor (i.e., one IV) Factorial – More than one factor (i.e., more than one IV) One-Way ANOVA When to use… • Interval/Ratio Data • One factor (i.e., IV) with two or more levels. – i.e., +2 means • Independent Observations

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 Expected Distribution • Null Hypothesis: the means are equal • Research Hypothesis: the means are not equal 01 2 3 :. . . 1 k HM M M M or F === 01 2 3 : ... 1 k H MMM M OR F ≠≠≠ ≠≈ F CRIT Distributions variance (differences) between sample means F= variance (differences) expected w/no tx effect Assumptions 1. Independence of Observations within and across groups random assignment is KEY 2. Normality Assumption •n > 30 minor violations are OK robust test like t-test so minor violations are OK. very small n’s might cause some alarm especially if you’re dealing with unequal n’s Assumptions 3. Homogeneity of Variance Across all groups Run F MAX test If reject the null, then violate the assumption. largest variance smallest variance MAX F = Note. The Hartley’s F-Max test assumes samples are all the same size.
3 Example s 2 = 8.143 s 2 = 9.2857 s 2 = 7.214 largest variance smallest variance 9.2857 7.214 1.2866 MAX F F = = = So …what now? what now? df numerator = k where k = number of levels for that factor df denominator = n-1 where n is the size of each sample • Our observed F MAX = 1.2866 compared to the critical value = ?? • Our table shows (n-1) for 12 and 15. What now? – Select 15 (i.e., the larger number) to be more conservative – This will give you a SMALLER F CRIT • More difficult to RETAIN the null. • Want to retain the null = more conservative! Example • Homogeneity of Variance? – YES – our observed F MAX value falls within the retain-region, so we retain the null. – There is no difference between the variances = met assumption! 3.54 4.90 1.2866 Homogeneity of Variance: SPSS • SPSS will give you the Levene’s test • Same logic for interpreting the results. – Retain the null (i.e., p > 0.05): Meet the assumption – Reject the null (i.e., p < 0.05): Violate the assumption Test of Homogeneity of Variances overal _sat .185 2 27 .832 Levene Statistic df1 df2 Sig.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 One-Way ANOVA • Two methods to use – General Linear Model (GLM) • Longer method and can use for unequal and equal n’s – Using descriptive statistics • Quicker method and can only use for equal n’s Method #1: GLM Handout: ANOVA (Independent Groups) • Question: Overall, how satisfied were you with your experiment experience?
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 17

OneWayANOVA_SP08_BB_nocolor - Test Guide 2 Groups...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online