Inferential Statistics
- Created by: KesiaKate
- Created on: 19-05-17 15:07
Why would you use a parametric test?
Population drawn from should be normally distributed, variances of populations should be approximately equal, should have at least interval or ratio data, should be no extreme scores
1 of 35
Why would you use non-parametric tests?
When assumptions of non-parametric tests cannot be fulfilled, when distributions are non-normal
2 of 35
What are the types of non-parametric tests?
Mann-Whitney U, Wilcoxon, Chi Square, Binominal Sign, Spearman's Rho
3 of 35
What aspects do you need to bear in mind when choosing a non-parametric test?
Type of data, experimental design, differences in conditions, looking for a relationship between co-variables
4 of 35
What is the observed value?
the number produced after the various steps and calculations for a statistical test have been carried out
5 of 35
What is the critical value?
a value taken from a statistical test table, which must be reached in order for results to be significant
6 of 35
What is the checklist for Chi-Square?
DV produces nominal type of data, independent measures design, exploring a difference between each condition (levels of the IV), or an association
7 of 35
Whats the first step of Chi Square?
Draw a contingency/frequency table and fill in the totals
8 of 35
What is the second step of Chi Square?
the expected frequencies for each cell needs to be calculated - Expected Frequencies = row total X column total / overall total
9 of 35
What is the third step of Chi Square?
Calculating Chi Square X ² (observed value chi)= Σ(sigma - sum of) (observed - expected)² / expected
10 of 35
What is the fourth step of Chi Square?
Add all the values in the final column - this will equal the observed value
11 of 35
What is the fifth step of Chi Square?
Calculate the degrees of freedom? (df) = (number of rows - 1) X (number of columns - 1) (REMEMBER: don't include total rows and columns)
12 of 35
What is the sixth step of Chi Square?
Using the table of critical values of chi square, use the degrees of freedom and the levels of significant (0.05 unless told otherwise)
13 of 35
What is the seventh step of Chi Square?
Compare the critical value and the observed value to establish if there is a significant result
14 of 35
How do you know if it is significant for Chi Square?
rule of R - the final observed Chi Squared value must be greater or equal to the critical value of Chi Square
15 of 35
What is the checklist for the Mann Whitney U test?
DV produces ordinal or interval type of data, independent measures design, exploring a different between each conditions (levels of the IV)
16 of 35
What is the first step of Mann Whitney U?
Rank (put in order from lowest to highest number) all scores together, rank all the different levels of the IV together
17 of 35
What should you do if two or more participants had the same score?
assign all the ranks (i.e rank 1,2,3) and add them together, you will then divide this by how many there is (1+2/2 = 1.5)
18 of 35
What is the second step of Mann Whitney U?
Add up all of the ranks for the levels of the IV separately this will equal R1 = Rank 1 (i.e 3+4+1.5+7.5+1.5+5.5)
19 of 35
What is the third step of Mann Whitney U?
Use the formula R₁/₂ - n₁(n₁+1)/2 use the SMALLEST value from when you added up the conditions (n = number of scores or participants)
20 of 35
What is the fourth step of Mann Whitney U?
Using the critical values table, use n₁ and n₂ to find the critical value
21 of 35
What is the fifth step of Mann Whitney U? Significance.
In order to determine whether there is a significant result, the observed value of U has to be equal or less than the critical value of U
22 of 35
What is the checklist for the Wilcoxon test?
DV produces ordinal or interval type of data, repeated measures design, exploring a difference between each condition (levels of the IV)
23 of 35
What is the first step of Wilcoxon?
To find the difference between each set of scores. Some of these differences may well be positives and negative differences.
24 of 35
What is the second step of Wilcoxon?
Rank the differences, ignoring any positive or negatives signs
25 of 35
What is the third step of Wilcoxon?
Look at the difference column and count how many positive figures and how many negative figures there are. Use the less frequent sign and add up the ranks and this is the observed value
26 of 35
What is the fourth step of Wilcoxon?
calculate n value, n=the number of differences (remember to take away any that were the same i.e had a difference of 0)
27 of 35
What is the fifth step of Wilcoxon?
Match the n value to the table of critical Wilcoxon value, should be at 0.05 significance
28 of 35
What is the sixth step of Wilcoxon? Significance?
In order to determine whether there is a significant result, the observed value has to be equal or less than the critical value
29 of 35
What is the checklist for Binominal Sign?
DV produces nominal type data, Repeated Measures design, Exploring a difference between each conditions (levels of the IV)
30 of 35
What is the first step of Binominal?
Note the flow of direction to see if its negative or positive, i.e if it goes yes to no = + whereas if no to yes = - and if its yes to yes or no to no then you IGNORE it
31 of 35
What is the second step of Binominal?
Count up each positive and negative sign and then choose the SMALLEST score and this is the observed value of S
32 of 35
What is the third step of Binomial?
Look at the critical values table to 0.05 and use n = number of participants (remember to exclude any people who were IGNORED)
33 of 35
How do you know if it is significant Binominal? Significant
In order to determine whether there is a significant result, the observed value has to be equal or less than the critical value
34 of 35
What is the checklist for Spearman Rho?
Appropriate for variables that produces at least ordinal type of data, Exploring a relationship between two co-variables, A correlation design has been used
35 of 35
Other cards in this set
Card 2
Front
Why would you use non-parametric tests?
Back
When assumptions of non-parametric tests cannot be fulfilled, when distributions are non-normal
Card 3
Front
What are the types of non-parametric tests?
Back
Card 4
Front
What aspects do you need to bear in mind when choosing a non-parametric test?
Back
Card 5
Front
What is the observed value?
Back
Similar Psychology resources:
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
5.0 / 5 based on 5 ratings
0.0 / 5
Comments
No comments have yet been made