Weeks 1-7 Flashcards
Describe what simple and main effects are?
simple effects are the statistical effects determined when analysing interaction effects, while main effects are the statistical effects of IVs
The SNK test is also known as….
Neuman-Kuells test
What is cluster sampling?`
Clusters are selected from the population, and then a random sample of individuals are taken from each of the selected clusters
What is Error Rate Per Comparison?
The probability of making a Type 1 error on any given comparison
In the case of two IVs we calculate three versions of _____, one for each IV and one for the interaction of the IVs treatment
MS
What two types of samples can you obtain?
Probability samples, and non-probability samples
Stratified Sampling?
The population is divided intro strata groups (subgroups) and then a random sample is selected from each of the strata
The ___ is heavliy influenced by extreme scores
What are mean, median, SD, mode called in a sample?
Statistics
A steep curve in kurtosis with a positive value is called
Leptokurtic
The difference between the largest and smallest value is called the ___
Range
Ó2 is the symbol for _____ variance
Population
How many F Values would you have if you had a factorial between groups ANOVA?
3: 2 IVs; 1 interaction
Is there always error in research?
Yes
What statistic do use receive from spss from ANOVAS
an F score
____ describes how many people there are in the extremes of the curve
Kurtosis
Compared to the one way ANOVA, in a one-way repeated measures ANOVA the ___ is reduced by removing the contribution made by individual differences
error
A ___ is when the effect of one IV on the DV depends on the level of the other IV
interaction
What is the third MS variable in a repeated measures ANOVA
MS between subjects
s2 is the symbol for ____ variance
Sample
The repeated measures ANOVA is very efficient so long as the scores _____ substantially
Correlate
The ‘test of within subjects effects’ table is the output in a …
One-Way Repeated-Measures ANOVA
In a mixed factorial ANOVA you ‘split’ the file by the ____-____ IV
between-subjects
If Ho is true, than MS error and MS treatment should be…
equal, or the same
When do you use a z test?
When you know the population standard deviation
What extra assumptions are their for some repeated measures analyses
Test of Sphericity
O 2 e is?
MS error
Generalisability of results, interaction effects, economy, and being a very useful, common design are all advantages of the …?
Mixed factorial design
Is a case study and survey method under the subgroup of descriptive research?
Yes
When looking at mixed factorial ANOVA interactions, what setting do you change in SPSS?
Split File
When would you use the Assumption of Homogeneity of Covariance?
When using a mixed factorial design
What does η2 Stand for?
Eta Squared
What does the Bonferroni correction calculate?
it adjusts the alpha level to equal alpha/ number of comparisons
What is a type 1 and 2 error?
1 when you incorrectly reject the H0, 2 when you incorrectly retain the HO
Is is most of the time easier to find a statistical difference within subjects?
Yes
When looking at mixed factorial ANOVA interactions, what setting do you change in SPSS?
Split File
Xij =μ+τj +eij What analysis uses this structual model?
ANOVA structual model
Sphericity is only useful when you have more than ___ level in your within-subjects variable with what test?
Mixed Factorial ANOVA
If MS treatment is a lot bigger than MS error than…?
there is a effect of the treatment, or manipulation
The ____ error is a common way of measuring sampling error?
standard error
To work out if there is a difference between several (more than two) groups, we calculate the population variance ____
Twice
______ of results is one of the major advantages of a Factorial Design
Generalisability
Homogeneity of Covariance is an assumption of the between subjects component of a …
Mixed factorial design
Xijk =μ+αi +βj +αβij +eijk What ANOVA uses this structural model?
Between Groups ANOVA
What 3 types of Factorial ANOVAS are there?
Factorial Between groups ANOVA, Factorial Repeated Measures ANOVA, Mixed (split-plot) Factorial ANOVA
The standard deviation of the sampling distribution of a statistic is called the ____ error
Standard Error
MS stands for?
Mean Square
When do you use a one-way repeated measures ANOVA?
When there is only one independent variable, more than three levels, and participants are tested twice
Descritpvie, predictive, and ____ research are the three types of Psychological research stratergies
explanatory
What is the Kurtosis name for : 1 a normal distribution, 2 a very steeped curve distribution, 3 a flat curve distribution?
1 mesokurtic, 2 leptokurtic, 3 platykurtic
Xijk* =μ+α*i* +π*j*/*i* +β*k* +αβ*ik* +πβ*jk*/*i* +*eijk
Mixed Factorial ANOVA
What is the inter quartile range?
The middle 2, 25% sections, (without the outliers or the outer 25% on each side)
Xij =μ+πi +τj +eij This is the structual model for what ANOVA?
Repeated measures ANOVA
What is convenience (grab) sampling?
Sample is drawn from an accessible population
What analyses do you use to follow up Factorial ANOVA interactions ?
Simple effect analyses
What is linear contrast>
Linear contrasts can be thought of as like a t test where you can compare not just groups but sets of groups (you assign weights to groups)
Is it impossible to attain all data from a population?
Yes
What are the most appropriate follow up tests to use for a One-Way Repeated Measures ANOVA?
repeated-measures tests using the Bonferonni correction
What are the two types of data?
Numerical and nominal
What is the most stable estimate of central tendency across samples?
Mean
What is Sphericity?
The variability in the differences between any pair of groups of scores should be approximately equal to the variability in the differences between any other pairs of group scores
Is numerical or nominal data quantitative?
Numerical
How many MSerror/s calculated in a between subjects factorial ANOVA?
1 MSerror
Explanatory research is usually involved in _____, while predicitve reserach is usually involed in ____
Experiments; correlation
Explain a 4x3x2x5 factorial design
4 IVs with levels of 4 3 2 5 respectively
Generalisability of Results, Interaction Effects and Economy are all advantages of …
Factorial ANOVA
The characteristics of a populations are called _____
Parameters
What is sample error?
Sampling error is the difference between sample data and population data (difference between statistics and parameters)
What is an inference?
A proposed causal link between our sample and how it represents a population
Name the three advantages of Factorial ANOVA?
Generalisability of results, Interaction Effects, and Economy
How many MSerror scores are calculated in a between subjects factorial ANOVA?
1
Power is…
The probability of correctly rejecting a galse null hypothesis
σ e2 = ∑ s 2j / k
Does this depend on the H0 Being True?
No it does not: this is MS error
What is Quota Sampling?
Population is broken into categories, then participants selected based on reaching a specified proportion
In the Mixed Factorial ANOVA, the Box Test of Equality is used for the Homogeneity of _____
Covariance
In the repeated measures ANOVA, to get homogenity of variance, you divide the largest sample variance by the smallest sample variance to get a F___
Fmax
You are more likely to have a ____ sampling error if you don’t use random sampling OR have a small sample
Large
Snowball Sampling?
Recruit participants based on recommendations from already recruited participants
What test do you use in regards to the normal distribution when you do not know the population standard deviation?
A t test
What is a type 1 and 2 error?
1 when you incorrectly reject the H0, 2 when you incorrectly retain the HO
Represents the largest number of people; Is a real score from the data set; can be used with nominal data…These are the advantages of what measures of central tendecny?
Mode
What is the Kurtosis name for : 1 a normal distribution, 2 a very steeped curve distribution, 3 a flat curve distribution?
1 mesokurtic, 2 leptokurtic, 3 platykurtic
______ effects are a very important advantage of Factorial Designs
Interaction
In skewness, is -.66 more to the positive or negative side?
Negative (hight to the right)
By calculating a statistic, we are giving a ____ estimate of a parameter, compared to a confidence interval which gives a ___ estimate of a parameter
point; Interval
When do you use a z test?
When you know the population standard deviation
Generalisability of results, Interaction effects, and Economy are the three advantages what ANOVA?
Factorial
It is always better to look at the effect of the within-subject variable at each level of the between subjects variable seperatly, in a ____ ____ ANOVA
Mixed Factoria ANOVA
What measure of central tendency is relatively unaffected by extreme scores?
Median
The best first thing to do when you have a significant interaction is ….
Look at the graph
σ e2 = n s X2
Does this depend on the H0 being true?
Yes it does: this is MS error
What are mean, median, SD, mode called in a population?
Parameters
What is a sample?
A set of scores from our population of interest
Xij* =μ+τ*j* +*eij
ANOVA structual Model
What is systematic sampling?`
Sample chosen from a list of all individuals in population by selecting every Kth individual
Xij* =μ+π*i* +τ*j* +*eij
Repeated Measures ANOVA
If the Fmax is less than ten, we can conclude….
That the homogeneity of variance assumption has not been violated
MSerror is the ___ condition variance, while MStreatment is the ____ condition variance
Within;Between
The most straightforward way of hypothesis testing of means is what type of test?
Z score
A flat curve that has a negative value is called
Platykurtic
It is easier to find statistical significance in a One-Way Repeated Measures ANOVA because over the two test sessions we can take out the constant ___ of people
the bit that is special to them
σ e2 = ∑ s 2j / k ?
MS error
The formula for F is?
MS treatment divided by MS error
Do we talk about RAW data when we are using a standard normal curve?
No, that is where we conceptually use Standard Deviation
To calculate _____ you pool the estimates of variance from each group in our study
MS error
There is (one/two) MSerrors in a mixed factorial design
Two; one for the between subjects, one for the within-subjects
what is Purposive sampling?
Selecting people from a particular category based on the focus of the research
A bimodal distribution means two ____
Modes
When do you use a Factorial ANOVA?
when there is more than one independent variable
What is an interaction?
when there is a significant statistical finding between an interaction between 2 IVs
Is is most of the time easier to find a statistical difference within subjects?
Yes
How many IVs does a one way ANOVA have?
1
Some of the error is due to ___ rather than the entire population
sample
Xijk* =μ+α*i* +β*j* +αβ*ij* +*eijk
Between groups factorial ANOVA
Does the MS error depend upon the HO being accurate?
No
Explain a 4x3x2x5 factorial design
4 IVs with levels of 4 3 2 5 respectively
What two types of samples can you obtain?
Probability samples, and non-probability samples
What is sample error?
Sampling error is the difference between sample data and population data (difference between statistics and parameters)
What is the square root of the data variance?
The Standard Deviation
When do you use Simple Effects analyses
ONLY when dealing with Factorial ANOVAs
What is the probability of correctly rejecting a false null hypothesis?
Power
What is MS error?
an estimate of the population variance
What is simple random sampling?
Every individual in the population has an equal chance of being part of the sample
Random assignment is used in ____ psychological research?
explanatory
The standard ____ is a common way of measuring sampling error
error
To calculate power you need to know…
Sample size, Alpha level, and desired effect size
The Games-Howl is the best post hoc test when what assumption has not been met?
Homogeneity of Variance
What is MS error?
an estimate of the population variance
When do you use Simple Effects analyses
ONLY when dealing with Factorial ANOVAs
σ e2 = n s X2
This is?
MS Treatment
Does the MS treatment depend assume that the population variance HO is true
yes
What does the πi in Xij =μ+πi +τj +eij ?
The constant of each individual
a = .05 is actually the __ rate per _____
Error; Comparison
Per comparison rate (a) x number of comparisons (c) = ca, or ….
the familywise error rate
What is the cutoff for the Box Test for Equality?
p
What is the standard error?
Is is the standard deviation of the sampling distribution of a statistic
What is the a main effect?
The influence of ONE IV separate of the others
What is a Familywise Error Rate?
The probability of when you are running multiple comparisons that it could contain at least one type 1 error
Is quasi-experimental designs explanatory or predictive?
Predictive
How many IVs does factorial ANOVAS have?
more than 1
What is a Familywise Error Rate?
The probability of when you are running multiple comparisons that it could contain at least one type 1 error
The post hoc test called ___ is really valuable for different samples with large variance
scheffe
What test has the assumption of homogeneity of Covariance?
Mixed Factorial ANOVA
If you have a problem with Sphericity in a one-way repeated measures ANOVA what ‘line’ should you use?
Huynh-Feldt
Xijk =μ+αi +πj/i +βk +αβik +πβjk/i +eijk What ANOVA uses this structural Model?
Mixed Factorial ANOVA
Most of the time in psychological we are working with what scale of measurement?
Interval
What is the standard error?
It is the standard deviation of the sampling distribution of a statistic (if you took many samples and grouped all their means)
What type of ANOVA are we using when it outputs Main Effects?
Factorial ANOVA
Are correlational and quasi-experimental predicitve or explanatory?
Predicitve
A kurtosis score of zero would be called
Mesokurtic
What is the Kurtosis name for : 1 a normal distribution, 2 a very steeped curve distribution, 3 a flat curve distribution?
1 mesokurtic, 2 leptokurtic, 3 platykurtic
What is the standard error?
It is the standard deviation of the sampling distribution of a statistic (if you took many samples and grouped all their means)
