Term 1 Revision Key words

Independent Variables
Variables that are manipulated
1 of 95
Dependent Variables
Variables that are measured
2 of 95
Quantitative variables
Variables that represent variation in amount
3 of 95
Qualitative variables
Variables that are in kind or type
4 of 95
Classification variables
Characteristics which are intrinsic to the subjects of the experiment
5 of 95
Nuisance variables
Potential independent variables which if left uncontrolled could exert a systematic influence on the different treatment conditions
6 of 95
Completely random design
Subjects are randomly assigned to serve in one of the treatment conditions: Between subjects design
7 of 95
Randomised block design
Blocks of subjects who are matched closely on some relevant characteristics
8 of 95
statistical hypothesis
A set of precise hypotheses about the parameters of the different treatment populations
9 of 95
Research hypothesis
Assumed nature of the world that gets translated into an experiment
10 of 95
Null hypothesis
Statistical hypothesis which will be tested
11 of 95
The Alternative Hypothesis
The values of the parameter in the different treatment populations are not equal
12 of 95
Alpha
Significant level of 0.05
13 of 95
Type 1 Error:
Reject the null hypothesis when the null hypothesis is actually true: False positive
14 of 95
Type 2 Error
Accept the null hypothesis when the null hypothesis is actually false: False Negative
15 of 95
What is ** total?
**within + ** Between
16 of 95
How do you work out Mean Squares?
**/DF
17 of 95
How do you work out between group Df?
DFA = a-1 (Number of groups - 1)
18 of 95
How do you work out within group Df?
DFa/A = AS-A (Number of pps - number of groups)
19 of 95
The Between group mean square
MSA = SSA/DFA
20 of 95
The within group mean squares
MSs/a = ** S/A / DF s/a
21 of 95
How do you work out the F ratio?
MSa/MSs/ a
22 of 95
When is F critical?
F observed has to be bigger than the F critical
23 of 95
How do you find a critical value?
Degress of freedom for the effect (A) and look along the horizontal axis of the F table, Degrees of freedom for the error term (S/A) and look down the vertical axis of the F table
24 of 95
What can the omnibus F ratio tell us?
That there is a difference between the means
25 of 95
What is a planned comparison?
Make predictions about the direction of the effects
26 of 95
What is a post hoc?
It is a test that is conducted after an experiment
27 of 95
How is a post hoc conducted?
Calculate the sum of the weighted means
28 of 95
What is the ** for planned comparison?
Number of subjects * Difference between compared means/ sum of coefficient which we weight the mean 2
29 of 95
How is the F ratio calculated?
To test the comparison: MSAcomp (Mean square of comparison = SSAcomp/DFcomp=SSAcomp/1
30 of 95
FAcomp
MSAcomp/MS S/a
31 of 95
Per comparison and familywise error
Performing multiple statistical tests on a set of data increases the chance of type 1 error
32 of 95
A type 1 error involving single tests are known as?
Per comparison errors
33 of 95
The whole set of type 1 errors is known as?
familywise error
34 of 95
What is the relationship between 2 error rates?
Afw = c (aPC)
35 of 95
Post hoc
Can't affor to ignore the familywise error rate
36 of 95
What are some post hoc tests?
Scheffe, Tukey HSD, T-tests --> All inccrease the chance of type 2 error
37 of 95
How to conduct the Scheffe test?
Fscheffe (a-1) F (DFa, DFS/A
38 of 95
What is a way to work it out?
N x Mean difference 2/2) / Mean squared within MSe
39 of 95
What is the Tukey test?
D= Q Square root (MSError/N)
40 of 95
What does the Tukey test work for?
Between group ANOVA with equal Cell sizes
41 of 95
What is the T-test?
When comparing two means, a modified t-test is available T=(mean 1- mean 2)/Square Root (2MSerror/n)
42 of 95
What is a T-test also known as?
Bonferroni correction test
43 of 95
What are the assumptions of the F ratio?
Independence, Random Sampling, Homogeneity of variance and normality
44 of 95
What is homogeneity of variance meaning?
The different treatment populations have the same variance
45 of 95
What is normality?
Observations are drawn from noramlly distributed populations
46 of 95
What is between group design?
Hartley's F max, Bartlett, Cochran's C
47 of 95
What is within/Mixed design ANOVA
Box's M
48 of 95
How do you test the homogeneity of Variance?
Largest Variance/Smallest Variance
49 of 95
How do you test Normality?
Skew, LilleFors, Shapiro Wilks
50 of 95
How do you measure out Z scores?
Skew-0/SeSkew
51 of 95
How do you work out the standard Error of the skew?
SEskew =Square root (6/N)
52 of 95
What does a Z score have to be?
Greater than 1.96 then sample is significantly different from normal
53 of 95
What is N?
The number of cases in the sample
54 of 95
What does a Skew do?
Reduces the probability of making a type 2 error
55 of 95
What do you do to Moderate, substantial or severe data that has a positive skew?
Square root, Logarithm, reciprocal
56 of 95
What do you do to moderate, substantial, severe that is negative skew?
square root, logarithm, reciprocal (K-X)
57 of 95
What is statistical power?
Probability of detecting real effect
58 of 95
What is the equation for statistical power?
Beta-1
59 of 95
What is Beta?
The probability of making Type II error
60 of 95
What does Statistical power depend on?
Alpha level, sample size , effect size
61 of 95
What is effect size?
Association between Dv and Iv, seperation of means relative to error variance
62 of 95
What happens when we make alpha less strict
We reduce to risk of type 2 error but increase type 1
63 of 95
What is sample size?
To small, an effect may be missed, too big and too expensive a study
64 of 95
What is Eta Squared?
Proportion of total variance that is attributed to an effect
65 of 95
What is partial eta squared?
Proportion of the effect and error variance that is attributable to the effect
66 of 95
R squared?
R2 is the proportion of variance explained by the model
67 of 95
What is Omega Squared?
Estimate of the dependent variable population variability accounted for by the IV
68 of 95
What is Cohens F?
One way between groups ANOVA
69 of 95
What are the 3 null hypothesis for an ANOVA?
The means of the different levels of the first IV will be the same, The means of the different levels of the second IV will be the same, the differences between the means of the different levels of the interaction will be the same
70 of 95
How is factorial ANOVA variability partitioned?
The effect of A=A-T, the effect of B= B-T, Interaction A x B}Known as residual
71 of 95
SSt
**a + **b +** ab +**sab
72 of 95
How is interaction mean squares?
MSAB = SSAB/DFAB where dfAB = (a-1)(b-1)
73 of 95
How is the error mean square?
MSsab = SSsab/dfsab
74 of 95
What is residual error?
Treatment Variable, subject variable
75 of 95
What is ANOVA assumptions?
Data from interval or ratio scale, normal distributions, independence, homogeneity of variance, sphericity,
76 of 95
Sphericity?
homogeneity of treatment difference variance
77 of 95
Mixed or split design
One or more repeated measures, one or more between subject measures
78 of 95
What is effect of interest?
A between group effect is estimated, a within subject effect is estimated, An interaction effect is estimated
79 of 95
What is the overall for Homogeneity of Variance?
Box's M
80 of 95
What is Within Subjects?
Mauchley's W
81 of 95
Between groups homogeneity of variance?
Levene's test
82 of 95
What are the three basic component of an ANCOVA
Effect, Error, Covariate
83 of 95
How is the treatment effect estimated?
When covariate scores are available we have information about difference between treatment groups that existed before the experiment was performed
84 of 95
What does ANCOVA use?
Linear regression to estimate the size of treatment effects given the covariate information
85 of 95
What can the adjustment for group differences can either what?
Increase or decrease depending on the dependent variables relationship with the covariate
86 of 95
What is the error variability in ANCOVA?
In between group ANOVA the error variability comes from the subject within group deviation from the mean of the group
87 of 95
What are the assumptions of an ANCOVA?
Normality of treatment levels, independence of variance estimates, homogeneity of variance, random sampling, linear regression, homogeneity of regression coefficient
88 of 95
Assumptions of linear Regression
This states that the deviations from the regression equation across the different levels of the independent variable have normal distribution with means of Zero, and heteroscedasity
89 of 95
What is homogeneity of regression coefficients?
The regression coefficients for each of the groups in the independent variable should the same
90 of 95
MANOVA
Perform an ANOVA style analysis on several DVs simultaneously
91 of 95
What is the assumptions of a MANOVA?
Multivariate normality, homogeneity of variance and singularity
92 of 95
What is the Discriminant function analysis?
Predict membership of groups
93 of 95
When is it used?
Groups are already known and the researcher is trying to find out what the differences are between the groups
94 of 95
What is the basic principle used in DFA?
Different functions are calculated that maximise the ability to predict membership of groups, the macimum number of functions calculated either the number of levels of the grouping variable less one, the number of degrees of freedom of the IV
95 of 95

Other cards in this set

Card 2

Front

Dependent Variables

Back

Variables that are measured

Card 3

Front

Quantitative variables

Back

Preview of the front of card 3

Card 4

Front

Qualitative variables

Back

Preview of the front of card 4

Card 5

Front

Classification variables

Back

Preview of the front of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Statistics resources »