Week 2: Standardisation

Question 1

Choose two methods of measuring behaviour and/or psychological traits/states. Outline the advantages and disadvantages of each.

Answer 1

Two methods of measuring behaviour/traits/states are using observational measures and questionnaires/interviews. The advantages of using observational measures include, it measuring actual behaviour/detailed information, and the fact that those being measured are unaware they are being monitored (less opportunity to change behaviour). The disadvantages include the level of detail of information can be limited depending on the experimental procedure and design, and that experimental control can be difficult. The advantages of questionnaires/interviews include very good experimental control, having detailed information collected easily, and it being both an easy and cheap method for collecting large samples of data. The disadvantages include it needing to be validated, it can be harder to use dynamic, complex stimuli, and that it sometimes relies on participant insight and honesty (which isn’t really reliable – for example self-reporting).

Question 2

Define psychometrics and psychological test.

Answer 2

Psychometrics is the science of psychological measurement. Psychological tests are a measuring device or procedure designed to measure psychology-related variables.

Question 3

Describe different categories of psychological test, giving examples of each.

Answer 3

There are two main categories of psychological tests – personality-type tests and ability tests. Personality-type tests are tests with no correct answer – there is no definitive correct answer. Ability tests are measuring actual ability, whether it can be done or not. Ability tests can be further split into achievement tests – tests of previous learning (e.g. educational tests) – and aptitude tests – tests of general ability and potential for learning (e.g. IQ tests).

Question 4

Where at UQ can you find more information about psychological and educational tests specifically (full URL not required)?

Answer 4

The UQ library website – they have pages about tests.

Question 5

Describe how you might go about finding information on a particular psychological test.

Answer 5

Through the Mental Measurements Yearbook.

Question 6

List five sources of information on psychological tests.

Answer 6

ETS test collection, books about individual tests or textbooks on psychological testing (through UQ library), academic journals, or publishers catalogues.

Question 7

What’s the point of standardising a score?

Answer 7

To help us interpret a score. Raw scores are difficult to interpret and compare.

Question 8

Give an example of a situation where it would be useful to standardise a score.

Answer 8

Height – height for a baby v adult would be different -> need to standardise to compare

Question 9

What is age norming?

Answer 9

Age norming is when a sample of people come from a certain age group (e.g. sample of 3-year olds).

Question 10

Give three examples of typical populations we might use as the basis of standardisation samples for psychological tests.

Answer 10

University students, drivers, school students

Question 11

What’s the advantage of having a big standardisation sample?

Answer 11

Bigger samples give a more accurate representation of the population and are more stable (less affected by sampling error).

Question 12

What’s the Law of Large Numbers?

Answer 12

The Law of Large Numbers is the fact that as a sample size grows, its mean gets closer to the true average of the population.

Question 13

Give some examples of the sort of populations that could make up a standardisation sample under different contexts.

Answer 13

Mental health in Aboriginal and Torres Strait Islander peoples – sample made up of Aboriginal and Torres Straight Islander peoples. Loneliness levels in uni students – sample made up of uni students. New therapy targeting bipolar I – sample consisting of diagnosed bipolar I.

Question 14

Discuss the issues we might need to consider when recruiting a standardisation sample.

Answer 14

What the context is of the design, if random sampling is able to be done correctly.

Question 15

What is “stratified sampling” and when might we want to use it?

Answer 15

Stratified sampling is where you deliberately recruit to get particular ratios of subgroups. It may be wanted to be used when achieving true random sampling is difficult.

Question 16

What strategies could you use to get a representative sample?

Answer 16

Random sampling or stratified sampling (if random sampling is difficult).

Question 17

How do you calculate a z score in Microsoft Excel?

Answer 17

(Raw score – AVERAGE) / STDEV

Question 18

How do you calculate a percentile rank in Microsoft Excel?

Answer 18

PERCENTRANK function

Question 19

What are the advantages and disadvantages of using percentile ranks?

Answer 19

The advantages are that it is easy to understand for laypeople, and it is easy to calculate. The disadvantages include the confusion between percentile rank and percentage correct, and that percentile ranks have a non-linear relationship with raw scores (differences between percentile ranks may have different interpretations depending on where you are on the scale.

Question 20

How do you calculate a T score in Microsoft Excel?

Answer 20

Calculate z score first, then multiple z score by 10, and add 50.

Question 21

An adult completes an intelligence test and their raw score is converted into a z score of -1. If their raw score were instead converted into an IQ score, what would it be?

Answer 21

IQ of 85

Question 22

How do you calculate an IQ score in Microsoft Excel?

Answer 22

Calculate z score, and then multiple the z score by 15 and add 100

Question 23

If a child scores 7 on an age-normed test that uses a stanine scale (where higher scores = better reading), then what does this mean?

Answer 23

The child would be in the 78-89 percentile rank, with 12% of people in their stanine.

Question 24

What are the properties of the stanine scale?

Answer 24

Stanine scales have 9 divisions, each based on .5 SD wide.

Question 25

How could you assign stanines easily in Microsoft Excel when you have hundreds of people in your sample?

Answer 25

Compute the percentile rank. Then select the data and sort by percentile rank. Stanines in order can then be entered manually.

Question 26

True or False: Previous learning can be described as aptitude.

Answer 26

False – previous learning can be described as achievement

Question 27

True or False: Aptitude and achievement tests are both types of personality test.

Answer 27

False – they are types of ability tests

Question 28

True or False: In a normal distribution, percentile bands occupy a smaller range of raw scores closer to the mean.

Answer 28

True – more people stacked up on each other -> more people with same scores, therefore each percentile band of people take up less space

Question 28

True or False: The raw score is the score that is yielded by a test before standardisation.

Answer 28

True

Question 29

True or False: In psychology, we generally need a normative sample in order to standardise raw scores.

Answer 29

True

Question 30

True or False: If we assume a normal distribution, then the number of people between the 50th and 55th percentile rank is the same as between the 90th and 95th percentile rank.

Answer 30

True – in both cases it is 5% of people

Question 31

True or False: If I convert the scores from two variables into T scores, then the magnitude of the Pearson correlation between the two variables will change.

Answer 31

False – T scores are a linear transformation, and do not affect the outcome of statistics such as correlations and t-tests

Question 32

True or False: A nine-year-old boy completes a test of reading comprehension and his raw score is converted into a z score of +0.95 compared with other children his own age (where a higher score = better comprehension). This means he is performing within the middle 68% of children his age, assuming a normal distribution.

Answer 32

True

Question 33

True or False: The number of people from a representative standardisation sample who would score 6 on a test using a stanine scale will be the same as the number of people scoring 8.

Answer 33

False

Question 34

True or False: