Flashcards
MCQ 2 (week 8-10)
Cards in this deck (14)
- 1for parametric tests:data must be continuous (interval or ratio)
scores must be a normal distribution
different samples must have the same variances - 2continuous dependent variablealso known as scale data, includes interval and ratio data
also includes ordinal data (data with categories such as 1st, 2nd, 3rd) - 3non-parametric dependent variablescores are not normally distributed
scores have ordinal information
different samples have different variances - 4two categorical IV'swithin (same) subjects = wilcoxon's t
between (different) = mann-whitney u - 5categorical dependent variablealso known as nominal or dichotomous when there are two variables
can use chi-squared - 6dealing with error variability randomisationessential in randomised control trials
can be used in both between and within subject designs
can use computer randomisation or a random number sequence - 7dealing with error variability placeboa placebo is a substance or treatment with no therapeutic value
- 8significance testingis p is less than 0.05 = significant
is p is more than 0.05 = not significant - 9when to use chi-squaredcategorical dependent variable
between subjects design
uses observed and expected count data - 10calculate chi-squared1. calculate row, column and overall totals
2. calculate expected values
3. compare expected and observed values
4. estimate probability and significance either using spss or the table which tells you the critical value depending on level of significance - 11reporting chi-squarede.g. a significant chi-squared of 43.91, 1 degree of freedom and 164 participants, we write:
x² (1,164) = 43.91, p < .0001 - 12calculating t-scoresis the t-score is far to 0, we consider the difference to be significantly different from 0 and reject the null hypothesis and accept the alternative
- 13p valuestells us how likely it is that our results are due to error variability or chance
if probability is low (p < 0.05) then we reject the null hypothesis - 14confidence intervalsgives us a range of values that the true population is likely to lie within
we compare two means
Information
- Description
- UNIVERSITY (USING DATA) Flashcards on MCQ 2 (week 8-10), created by Aamna Khan on 04/04/2019.
- Created by
- Aamna Khan
- Views
- 1
- Rating