# Levels of Measurement

?
• Created by: FatCat3
• Created on: 01-03-23 20:52
what tests do you have to know how to calculate?
the sign test is the only one you need to know how to calculate, all other statistical tests dont need to be calculated
1 of 19
whats the 3 ds?
1. difference
2. design
3. data type
2 of 19
what does difference refer to?
if its testing for aa difference of a correlation
3 of 19
what does a test of difference include?
.measuring the impact of the IVs on the DV
.research effects difference between groups
.investigation of cause and effect
.experimental methods used to test difference
4 of 19
what does a test of correlation include?
.testing the rs of covariables (if one effects the other) rs between variables investigated
.investigation of an association
.any method can be used, it is just analysed differently
5 of 19
what does design refer to?
if researchers use an independent groups design, repeated measures design or matched pairs (recall each design and its +/-)
6 of 19
when is the design feature needed to be uses?
only when testing for a difference, not an association, if an association is being tested they can move on to the next level of measurement
7 of 19
how can the design help researchers?
it allows researchers to understand the type of data that they should be used
which will be either related or unrelated
8 of 19
what does related data mean?
related refers to data in which pp in each condition are related in some manner meaning repeated measures or matched pairs deign was used
9 of 19
what does unrelated data mean?
unrelate refers to data that uses 2 separate groups of ppl in each condition of the study- independent groups design was used
10 of 19
what are the levels of measurements?
(NOI)
nominal
ordinal
interval
11 of 19
what is nominal data?
refers to categorical data ie if a researcher was interested to know if more students doing A level psychology went to school or college, data would be categorised as either school or college, there are 2 distinct categories, pp can only appear in 1 categ
12 of 19
name the evaluations
+typically generated from closed questionnaires so is easy to generate quickly
-no scale of reference for nominal data, its categorised so doesnt express true complexity- oversimplified
13 of 19
what is ordinal data?
when data is ordered in the same way, intervals between data are not equal, typically used to simply rank data where the values attached have no meaning beyond the purpose of stating where one score appears to others
14 of 19
what are examples of this data that can be researched?
.researchers typically investigate non-physical entity such as attitudes
15 of 19
name the evaluations
+provides more detail then nominal as the scores are ordered in a linear fashion
-intervals between scores aren't equal, may be subjective, taking away objectivity
16 of 19
what is interval data?
like ordinal data as it fired to data that is ordered in the same way but with intervals, researchers are confident that the intervals between each value are equal in measurement
17 of 19
what can be examples?
eg time and temperature- the 3-4* is the same as 38-39*
18 of 19
name the evaluations
+more objective/scientific in nature and reliable
-intervals can be arbitrary ie 100* is not twice as warm as 50*, can only state the diff between intervals
19 of 19

## Other cards in this set

whats the 3 ds?

1. difference
2. design
3. data type

### Card 3

#### Front

what does difference refer to?

### Card 4

#### Front

what does a test of difference include?

### Card 5

#### Front

what does a test of correlation include?