inferential stats

Question 1

what are inferential statistics?

Answer 1

they help us infer meaning from the data ie they help draw conclusions and make predictions

Question 2

why does psychology use statistical tests?

Answer 2

to test the probability that an event occurs due to chance

Question 3

what does p ≤ 0.05 mean?

Answer 3

means that there is a 5% chance that the results happened due to change

it is the standard p value in psychology

Question 4

what does p ≤ 0.01 mean?

Answer 4

there is a 1% chance that the results happened due to chance

you may decrease the p value for new medicine tests etc

Question 5

what is a type 1 error?

Answer 5

if you accept your alternate hypothesis when you should have accepted the null it is a type 1 error

is a false positive

Question 6

what is a type 2 error?

Answer 6

if you accept your null hypothesis when you should have accepted the alternate hypothesis it is a type 2 error

is a false negative

Question 7

when do you use a chi-squared test?

Answer 7

when the data is nominal and it is an independent measures design

Question 8

when do you use a mann-whitney U test?

Answer 8

when the data is ordinal or interval and the design is independent measures

Question 9

when do you use a binomial sign test?

Answer 9

when the data is nominal and the design is repeated measures/matched pairs

Question 10

when do you use Wilcoxon Signed Ranks test?

Answer 10

when the data is ordinal/interval and the design is Repeated Measures/ Matched Pairs

Question 11

when do you use a Spearman’s Rho test?

Answer 11

when you have a correlation (with ordinal/interval data)

Question 12

why would you use a parametric test?

Answer 12

if your data is normally distributed

Question 13

when do you use a non-parametric test?

Answer 13

when your data is skewed

Question 14

what is your observed value?

Answer 14

is the value from your data

Question 15

what is a critical value?

Answer 15

found in a critical values table.

you can compare your observed value to this to see if your result is significant

Question 16

in which stats tests does the observed/calculated value need to be less than the critical value to be significant?

Answer 16

binomial sign test

wilcoxon signed ranks

mann whitney u

Question 17

in which stats tests does the observed value need to be more than the critical value to be significant?

Answer 17

chi squared

spearmans rho

Question 18

what are the calculation steps for mann whitney u?

Answer 18

• rank scores (from lowest to highest)
• use formula to find U:

U1 = n1n2 + n1(n1+1)/2 – R1

U2 = n1n2 + n2(n2+1)/2 – R2

• calculate the smaller U
• if the calculated value U is less than the critical U value, reject null
• if calculated U value is higher than critical U value, accept null

Question 19

what are the calculation steps for wilcoxon signed ranks test?

Answer 19

the difference in values of each condition is calculated, differences are then ranked.

the sum of the positive and negative values is found and T is the smallest of these values

if the calculated W is less than the critical value, reject null

if the calculated is higher than the critical, accept null

Question 20

what are the calculation steps of chi-squared test?

(note: practice this don’t just rely on the flashcard for the love of god)

Answer 20

add totals for each column

calculate observed and expected frequencies (use formula):

χ2 = Σ(Oi − Ei)^2/Ei

chi value - adding all cells of expected frequencies

if the calculated Chi value is less than the critical Chi value, accept null

if the calculated Chi value is higher than critical Chi value, reject null

Question 21

what are the calculation steps for binomial sign test?

Answer 21

determine positive and negative values for data

add each positive and negative assigned direction

smallest direction score = S value

if the calculated S value is less than critical value, reject null

if calculated is higher than critical value, accept null

Question 22

what are the calculation steps for spearman’s Rho correlation?

Answer 22

rank each dataset individually

find difference between each dataset

square difference (d2) and find the total of (d2) column

use formula to find rho:
r = 1 - (6ΣD^2)/(n(n^2-1))

just search up the formula i got no clue how to write it out on these flashcards and im like 90% sure you don’t even need to memorise it

if the calculated Rho value is less than the critical value, accept null

if calculated is higher than critical, reject null

Question 23

how to calculate the degrees of freedom?

Answer 23

(number of rows of data - 1) x (number of columns of data -1)