Flashcards in Statistics Deck (30):
what are descriptive statistics?
they describe samples of the population in terms of numbers
they include means, medians, SD, Z-scores and Percentages etc
Generally descriptive statistics should be given to 2 D.P.
what are inferential statistics?
they infer meaning from our samples.
Include Pearson's R, the Mann Witney test and the T-test.
what is interval data?
this is data with similar intervals between them (like a ruler)
e.g calorie intake
What is ratio data?
ratio is very similar to interval data except it has an absolute 0 value.
e.g a memory test.
what is Variance?
It is the measure of average deviation of the scores from the mean score. The bigger the variance the more spread there is around the mean
What is the interquartile range?
this is the range of scores of the middle 50%
what are Z scores?
a z score is probability of density function.
a Z score will tell you how many standard deviations your score is from the mean
they can be negative or positive
what are the parametric assumptions?
1. it needs to be interval/ ratio level data
2. data needs to be normally distributed
3. homogeneity of variance/ sphericity ect
what is homogeneity of variance?
The assumption of homogeneity of variance is that all comparison groups have the same variance.
What is sphericity ?
Sphericity applies to a repeated measures design. It is the condition where the variances of the differences between all combinations of related groups (levels) are equal
What are Psychometric tests??
these are implicit tests for implicit psychological measures and unconscious aspects of human nature.
they are useful as they try to avoid response bias
Name a few types of psychometric tests.
1. Attitude questionnaires
2. word association tests
3. semantic differential tests (type of attitude scale)
4. implicit tests e.g IAT
what are the problems with attitude questionnaires?
1. the Normative response bias: people give answers that are considered right or normal e.g I always tell the truth
a lie test can be used to combat this
2. social desirability effect
3. Ambivalence- people can report both positively and negatively for the same attitude
Evaluate word association tests.
1. they are easy and cheap to administer
2. they can be used to predict drug use (Ames et al)
3. they can be self scored which improves validity
4. colloquialism- words that are drug related in one area may not be drug related in another
5. implicitness is questionable?
What is the IAT- implicit association task
it is used to measure association between concepts.
widely used in social research- e,g prejudice ( age race gender)
evaluate the IAT.
1. Suggested it measures cultural views rather than personal beliefs
2. lacks ecological validity- cannot be generalised to real life.
what are correlations?
a correlation is a relationship between two variables..
they are specific models that allow us to make inferences about the population
what is the difference between Spearman's and Pearson's correlations
Pearson's R correlations only look at linear data and are very powerful
Spearman's Correlations can look at non-linear data
what is the equation of a straight line?
y= bx + a
b is the gradient of the line
a is the intercept
what is a model?
a model is made of data and some form of natural function that can explain the outcome
e.g a model can be Pearson's r correlation coefficient or a regression line
what does Pearson's R particularly assume?
the assumption of Linearity.
what is Cronbacks alpha?
this is a test that splits items equally in every way possible
it then correlated all halves with all other halves
it is a very robust measure of internal reliability.
what is convergent validity ?
this is comparing your test to a similar test- A HIGH CORRELATION INDICATED VALIDITY.
what is divergent validity?
correlating your study to an unrelated construct.
if you are looking to measure depression, a high correlation with an anxiety questionnaire would make your depression scale less valid.
What is a type 1 error?
When the null hypothesis is rejected when it should be accepted
What is a type 2 error?
This is when the null hypothesis is accepted when it should be rejected
What is interpolation?
Reading off values of y from any given value of x
What is extrapolation
We extrapolate data when it goes on past the plotted data values
Summarise and Evaluate regression models..
They have a predictive utility. They can be used to look at multiple predictor variables.