This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: To include results in the report Descriptive Descriptives aims to reflect the pattern of responses found in the sample collected...the simple stuff (mean, mode etc). Remember to include percentages. Look at valid percent Straight forward, interpret data In the report, you can write 50% of population does this, 20% does that Crosstab with ChiSquared Hypothesis example: H : There is no association between gender and modem speed H 1 : There is an association between gender and modem speed Crosstabulation allows you to examine two (or more) variables, and is used to assess whether there is an association/relationship between variables. Tests the presence of an association between 2 nominal or ordinal variables Reference percentages % not means The Chisquare test allows you to assess whether group or category membership on one variable is influenced by group or membership on another variable. Again, like descriptives, simple and p ercentag es However, significance (Pearson ChiSquared line Sig. 2 sided) m ust b e < 0.05. higher, still m e ntion the p ercentag es but say its not significant, and will not re further a n alysis Look at bottom textline. The % in brackets must be lower than 20%. If it is higher, mention that the results might be invalid and might require recoding. 20% of males did this, 30% of females did that compare One sample ttest Used when there is data from a single sample of participants, to test assumptions about whether the mean of the population is the same as the hypothesised mean To describe how the entire population thinks/behaves/responds with regards to one variable. First time hypothesis is introduced. Include it in report, just state if null is reject (never accept) In order to determine wheth er to reject/not reject null, we look at t stat OR pva look at pvalue).  If p value is < 0.05 (its significant), we DONT reject This what bigger than that (quote m e a ns), a nd/but it was/wasnt significance (quote pvalu reject/dont reject null (state null) Independent ttest Hypothesis example Ho Null Hypothesis: on average, there is no difference in the importance males and females place on characteristics of an ISP H1 alternate hypothesis: on average, there is a difference in the importance males and females place on characteristics of an ISP Similar to onesample, except youre now comparing a preference/behaviour between TWO (or more) different groups. If objective asks for how do x and y differ in their... Remember: Independent for 2 variables (gender), ANOVA for 3 (age) Paired ttest Same as Onesample/independent, except now you compare how the entire sample differ in two variables (difference to the responses to two separate questions)...
View Full
Document
 Three '10
 mark

Click to edit the document details