Statistical testing Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

Unrelated design (Independant design)

C
M
U

A
  • Chi-squared
  • Mann-whitney
  • Unrelated T-Test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Related Design (Repeated measures)

S
W
R

A
  • Sign Test
  • Wilcoxon
  • Related T-test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Test of association or Correlation
C
S
P

A

Chi-squared
Spearmans’s Rho
Pearson’s R

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Ordinal data
example mean

M
W
S

A

Mann-whitney
Wilcoxon
Spearman’s Rho

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Nominal Data

A

Chi-squared

Sign Test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Interval data

U
R
P

A

Unrealted T-Test
Related T-Test
Pearson’s R

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Mnemoic

Carrots should come Mashed With Swede Under Roast Potatoes

A

N =Carrots Should Come
O = Mashed With Swede
I = Under Roast Potatoes

Unrelated design

  • Chi-squared
  • Man whitney
  • Unrelated T-test

Related design

  • Sign test
  • Wilcoxon
  • Related

Test of association or correlation

  • Chi-squared
  • Spearmans rho
  • Pearson’s R
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Type I error

A

If significance level is too lenient e.g 0.1 (10%) it can result in the null hypothesis being rejected when it is the case (optimistic error or a false positive) as the researcher claims to have found a significant difference/correlation when one does not exist.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

A Type II error

A

If significance level is too stringent e.g. 0.01 (1%), it is possible that is the null hypothesis will be accepted when in fact it is false (pessimistic error or a false negative).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Define what is meant by the critical value

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Step one of sign test

A

State the hypothesis.

  • If hypothesis is directional = One tailed test
  • If hypothesis is non-directional = Two tailed test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Step two of sign test

A

Record the data and workout the sign where the pair of data is (+) or (-)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Find calculated value

A

S is found by calculated by adding up the pulses and adding up the minuses and selecting the smaller value.
S= calculate value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Step 4 of sign test

A

Find cirtical value of S
- N = total number of scores (Ignore all zeros)
Then use table of critical values and locate column 0.05

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Step 5 of the sign test

A

If the hypothesis is directional check that the result is expected direction

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is nominal data?

A

Nominal level data is frequency or count data that consists of the number of participants falling into categorie

17
Q

what is ordinal data?

A

data which is placed into some kind of order or scale.

18
Q

What is interval data?

A

data measured in fixed units with equal distance between points on the scale.

19
Q

if the data is nominal and independent what test is used?

A

chi-squared

20
Q

if the data is nominal and repeated what test should be used?

A

isign test

21
Q

if the test is ordinal and testing for correlation?

A

it is spearman’s rho

22
Q

if the rest is testing for difference and repeated measured what test should be used

A

Wilcoxon

23
Q

if the data is ordinal and independent groups what test should we use?

A

Mann-whitney

24
Q

If the test is interval/parametric and has a correlation what test would you use?

A

Pearson’s r

25
Q

If the test is interval/parametric and has a test of difference and repeated measures design?

A

Related t-test

26
Q

If the test is interval/parametric and has a independent groups design?

A

un-related t-test