psychology comp 1: inferential stats

inferential stats

what is inferential stats?

helps researchers decide by working out the probability of the difference between the two sets of scores occurring just due to chance- this probability is expressed as a number between 0.00 (0% chance) and 1.00 (100% chance)

what is a type 1 error?

it is a false positive that rejects the null hypothesis and accepts the alternate hypothesis although really the null hypothesis was true - significance level is too lenient, too positive (not being cautious enough)

what is a type 2 error?

it is a false negative that rejects the alternate hypothesis and accepts the null hypothesis although really the alternate hypothesis was true- significance level is too strict, too negative (being over cautious)

what are parametric tests?

tests that rely on reasons and assumptions about the distribution

what are the 3 criterias for parametric tests?

1) should only be used on data of interval/ratio status

2) the data will come from a sample drawn from a normally distributed population

3) there is a similar variance between conditions

what is the observed value?

the number produced after the various steps and calculations for a statistical test have been carried out

what is the critical value?

a value taken from a statistical test table which must be reached out in order for results to be significant

what is the rule about inferential statistic tests?

if the test has an R in its name the observed value must be greater than the critical value for it to be significant

what is the checklist for Mann Whitney U test?

DV produces ordinal or interval level data

Independent measure design

exploring a difference between each condition (levels of the IV)

what is ordinal level data?

data that is placed into some kind of order or scale eg: rating happiness on a scale of 1-10

what is interval level data?

data that is measured in fixed units with equal distance between the points on the scale eg: temperature, scores etc

what is ratio level data?

similar to interval data it is measurements on a scale but it is set to absolute zero so there can’t be negative values eg: length measured by a ruler in centremeters starts at 0- no negative length

what is nominal level data?

frequency or data that is used to label variables without any quantitative value eg: hair colour, male or female etc (NOMinal- name)

what are the steps to calculating Mann Whitney U test?

step 1- rank all scores together, ignore which groups the ranks are associated to

rank the scores from lowest to highest

! if there are two or more of the same number calculate the rank by working out the mean of the ranks that are given !

step 2- add up all the ranks for A to get the value of R1 (rank 1)

step 3- repeat the same process as step 2 for B to get the value of R2 (rank 2)

step 4- formula use

what is the checklist for using the Wilcoxon sign test?

DV produces ordinal or interval level data

Repeated measures design

exploring a difference between each condition (levels of the IV)

what are the steps to calculating Wilcoxon sign test?

step 1- find the difference between each set of scores, these may be positive or negative

step 2- rank the scores from lowest to highest

! if there are two or more of the same number, calculate the rank by working out the mean of the ranks that would be given!

! if the difference is 0, omit the result and reduce N accordingly!

step 3- calculate the observed value by working out the sum of the ranks of at least frequent sign

step 4- find there critical value and compare it to the observed value