Please enable JavaScript.
Coggle requires JavaScript to display documents.
Descriptive Statistics - Coggle Diagram
Descriptive Statistics
-
Variables
Indépendant variables
- The variables we pre-set such as sex and the measurements of them, for example, 1=male 2=female
- Indépendant variable is a variable that is manipulated by a researcher to investigate whether is consequences brings change to another variable.
Dependant Variable
- Dependant variable, the variable whose measurement is dependant upon how respondents answer our question
- DV which is measured and predictable to dependent upon the IV, is therefore named the dependant variable
Levels of Data
Nominal
- Categories
- Labels
- Names
- Gender
-
-
Ordinal
- Ordered categories
- Scaling
- Place in a Race
Having a mathematical relationship
1 = Strongly agree
6 = Strongly Disagree
Where each data point is progressive and evenly spaced
-
Interval
- Can be measured but has no zero
- Temperature
-
Ratio
- Measurable and has true zero
- Height
-
Definitions
- N means number of participants
- Min and Max, Range of scores and the smallest and highest score
- Mean = the average score
- Std. Dev = Standard deviation, how much on average the individual values differ from the mean, the smaller the SD the less each score varies from the mean
- Mode, the most frequent value
- Median, middle value
- Skewness, this provides an indication of the symmetry of the distribution
- Kurtosis = Information on the peakedness of the distruption
Inferential Tests, Parametric or not?
Parametricity broadly centres around three assumptions critera
- Data is ordinal
- Normally distributed
- Homogeneity of variance
Homogeneity of Variance
- Variance is numerical how far a set of numbers is spread out from their average value, within a particular variable
- Homogeneity of variance is the similarity of variance in two or more variables , eg Groups male or females
- Can be calculated using SPSS, Levenes test for equality of variances
- You want non significant results
- The use of Rule of Five, the variance within one variable should not exceed five times that of the others