Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Semester 2 Psych > Statistics > Flashcards

Statistics Flashcards

You may prefer our related Brainscape-certified flashcards:
Psychology 101 flashcards GRE® Psychology flashcards
1
Q

Non-parametric examples

A

Mann-Whitney
Wilcoxon
Chi2
Sign test
Spearman’s rank

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

Parametric examples?

A

Unrelated t-test
Related t-test
Pearsons R

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

Independent / between subject design table

A

Normally distributed - independent samples t-test

Not normally distributed - Mann whitney

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

Repeated / within subject design table

A

Normally distributed - paired samples t-test
Not normally distributed - wilcoxon W

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

Parametric test - describe the data

A

Normally distributed
No extreme outliers
Data shouldn’t be correlated

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

Non-parametric tests - describe the data

A
  • not normally distributed
  • data is ordinal (set order/scale)
  • small sample
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is central tendency?

A

Summarisation of a data set with a single number - mean, media, mode

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is a distribution?

A

A set of data

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

What is a concept of variance?

A

Extent to which the data values are spread out from the mean - SD

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

What is SD related to?

A

Unsystematic variation

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

What is unsystematic variation?

A

It is uncontrolled - due to chance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is used to test the quality of the experimental result?

A

Systematic variation / unsystematic variation

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

What variance is desirable and what variance is undesirable?

A

Systematic variance is desirable
Unsystematic variance is undesirable - can obscure the systematic variance you are looking for

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

What is unsystematic variation?

A

Variation is not due to the effect we are interested in. This could be due to natural individuals differences

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

What is systematic variation

A

Variation due to a genuine effect or experimental manipulation

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

What are normal distributions?

A

Have similar mean, median and mode and the data is spread symmetrically on both sides - same amount above and below the mean

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What produces a normal distribution?

A

Standardised tests

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What do Z scores tell us?

A

It’s a way of talking about data in germs of number of standard deviations above and below the mean

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What do Z scores allow us to assess?

A

Where an individual falls relative to the population as a whole

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

How to calculate Z scores?

A

Your score - mean score / standard deviation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is the threshold for normally distributed data?

A

+-1.96

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

How many people fall into the +-1.96 range?

A

95%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What does the P value tell us?

A

The probability that the observed, or more extreme results could have occurred if the statement was true

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

6 stages of hypothesis testing?

A
  1. Set out research hypothesis
  2. Set out null hypothesis
  3. Assume null to be true
  4. Calculate Z scores
  5. Convert to p value
  6. If p<.05, reject null
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What does sampling distributions tell us?

A

How the sample means vary - this isn’t a distribution of actual scores but a mean value across many different samples (a distribution of sample scores)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is population distribution?

A

Variation measures in standard deviation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

How is sampling distribution measured?

A

Variation measured in standard error

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What is sampling distribution of the mean?

A

Has a smaller SD than the population mean, converges on the true mean
Looks different as N changes (the higher N is, the smaller the SD)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What is the standard deviation of a sampling distribution called?

A

Standard error

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Why is it good to be interested in sampling distribution?

A

We know the mean will vary, even if you repeatedly sample from the same population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

How you do work out T score?

A

Difference in means / standard error of difference

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What is the difference between t statistics and Z scores?

A

T statistics are comparing 2 groups - 2 sets of variabilities
Whereas Z scores are comparing 1 individual or small group to the whole population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

When going from T to P why is it important to know the degrees of freedom?

A

Need to know this because the shape of the T distribution changes as a function of how many subjects we have in the study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

What are independent group at tests also known as?

A

Unrelated t-tests
Unpaired means t-test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

What are paired means t-test also known as?

A

Related t-tests
Dependent t-tests

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

What happens if the data is not normally distributed?

A

Sample mean and standard deviation are no longer representative of the population - parametric test no longer gives accurate p-value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

What does it mean if data is positively skewed?

A

Mean is bigger than the median
The data’s high point falls to the left

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

What happens if the data is negatively skewed?

A

Mean is smaller than the median
High point falls to the right (NR)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

Positive skew falls to the left because…

A

It pulled the mean upwards

40
Q

Negative skew pulls to the right because…

A

It pulled the mean downwards

41
Q

What is leptokurtic Kurtosis?

A

Slender peak (tall and narrow)

42
Q

What is platykuryic?

A

Broad peak (flat, wide)

43
Q

What are 4 things to do with a non-normal distribution so we can analyse it?

A
  1. Transform it to make it normally distributed - reduces skew but breaks links between data and context
  2. Remove outliers - reduces skew but removes data
  3. Re sample data many times to create normal distribution
  4. Use non-parametric tests - with ratio/interval data, it’s changed to ordinal
44
Q

What is ratio data?

A

The scale has a 0 point - age, income

45
Q

What do parametric tests do?

A

Make many assumptions about the data

46
Q

What do non-parametric tests do?

A

Don’t make any assumptions about the data

47
Q

Mann Whitney test - what happens when the null hypothesis is true?

A

The scores of the 2 groups to be randomly distributed

48
Q

Mann whitney test - what would we expect if the null hypothesis was fake?

A

The scores of the 2 groups to be clustered at either end of the sequence

49
Q

What is a correlation?

A

The amount of association between 2 variables

50
Q

What does a correlation determine?

A

Whether the relationship is reliable

51
Q

What dies homoscedasticity mean?

A

Correlational data being equally distributed about the regression line

52
Q

What is the non-parametric correlation?

A

Spearman’s rank - this uses same method as pearson’s R but first converts the raw data into ranks

53
Q

What does a positive sign indicate in correlation?

A

Convergent association

54
Q

What does a negative sign indicate in correlation?

A

Divergent association

55
Q

What weakens spearman’s rank?

A

Tied ranks (ranks with the same number)

56
Q

When is Kendall’s rank correlation better to use?

A

Small samples (smaller than 12)

57
Q

What makes adjustments for tied ranks?

A

Kendall’s Tau-b

58
Q

What non parametric correlation is most commonly used?

A

Pearsons R - relies on strict assumptions and is very susceptible to outliers

59
Q

When is a questionnaire appropriate?

A

When wanting to know what people think, the content of their mind

60
Q

What to questionnaires probe?

A

Subjective experience to record the related conscious reaction

61
Q

4 advantages of using a questionnaire?

A
  1. Participants can fill out in their own time
  2. There are ways to test the reliability of the questionnaire
  3. Inexpensive
  4. Easier to test large numbers of participants
62
Q

5 disadvantages of objective measures?

A
  1. Experimenter has to be present
  2. Resources can be expensive
  3. Reliability is more difficult to test
  4. Financial costs are greater
  5. Takes longer to test in the lab, and harder to test and recruit large numbers of participants
63
Q

What is validity?

A

The extent to which questionnaire scores represent the variable they are intended to measure

64
Q

Name 5 forms of validity?

A
  1. Face validity
  2. Construct validity
  3. Convergent construct validity
  4. Divergent construct validity
  5. Content validity
65
Q

What is test retest reliability?

A

Seeing if the participants give consistent responses if they are repeatedly tested on the questionnaire

66
Q

What is inter-rater reliability?

A

Used when 2 or more raters make judgements or observations on another subject - reliability checked by comparing the scores given by the 2 raters for the same subject

67
Q

How to overcome language barriers in questionnaires so the meaning of the question isn’t lost?

A

Translate the question then translate the question back into the source language to check that they match the original worded questions

68
Q

What does back translation do?

A

Validates the quality of translated research questionnaires

69
Q

Advantages of having a researcher present?

A
  1. Expertise ( clarify questions, consistent with the goals of the questionnaire)
  2. Personal knowledge of the world
70
Q

Disadvantage of having a researcher present?

A

Can hinder experiment by bringing in their own beliefs, attitudes and expectations - might lead to tester effects

71
Q

What is a data extract?

A

Part of the data that is used to support your interpretation

72
Q

What is a data item?

A

One item within the dataset

73
Q

What is a dataset?

A

The collection of data items that have been complied to be analysed

74
Q

What is a code?

A

A label that you give to a proportion of the data which represents what you think a specific section of the data means

75
Q

What is a rationale?

A

A justification for your research

76
Q

Where should the rationale be?

A

Throughout the introduction

77
Q

In qualitative reports, what should the rationale do?

A

Also justify why a qualitative approach is relevant

78
Q

What is reflexivity?

A

A researcher critically reflecting on their values, assumptions, expectations, choices and actions throughout the research process and considering how these might affect analysis

OUR ASSUMPTIONS AND CHOICES INEVITABLY SHAPE THE KNOWLEDGE WE CREATE

79
Q

What does qualitative research aim to explore?

A

Actors constructions

80
Q

Who came up with the criteria for evaluating qualitative research?

A

Lincoln and Guba

81
Q

What is the criteria for qualitative research?

A
  1. Credibility
  2. Transferability
  3. Dependability
  4. Confirmability
  5. Authenticity
82
Q

What is authenticity?

A

Does the research represent a fair range of differing viewpoints on the topic

83
Q

What is credibility?

A

Do participants or members of the community being researched feel that the findings represent their experience?

84
Q

Qualitative research that is high in credibility may have what?

A

Prolonged engagement with the participants
Negative case analysis
Triangulation

85
Q

How can authenticity be assessed?

A

Researchers doing ‘member checking’

86
Q

How can credibility be assessed?

A

Member checking
Peer debriefing with other researchers

87
Q

What is transferability?

A

Are the findings applicable in other contexts?

88
Q

Qualitative research that is high in transferability may have…

A

Rich description of participant responses
Rich description of researcher interpretations

89
Q

Transferability can be assessed by?

A

The reader having enough contextual information and depth of understanding to be able to recognise this aligns with my experience

90
Q

What is dependability?

A

Would similar findings be produced if someone else undertook the research

91
Q

Qualitative research that is high in dependability may have…

A

Multiple researchers exploring the same research question
Transparency in regards to process of analysis

92
Q

Dependability can be assessed by

A

Peer debriefing

93
Q

What is confirmability?

A

Are the findings a product of the data and not the researchers biases

94
Q

Qualitative data that is high in confirmability may have…

A

Transparency in regard to process of analysis
Reflexivity

95
Q

Confirmability can be assessed via

A

Clear insight into the researchers experiences and position

Semester 2 Psych (5 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions