Task 4 Flashcards
What is an association?
βif the value of one variable tells you something about the value of the other variable, then they are associatedβ
Which 2 types of variables are there?
dependent variables
independent variable
What does the dependent (=response) variable measure?
measures outcome of study
What does the independent (=explanatory) variable measure?
causes/explains change in dependent variable
What does a Scatterplot show?
The relationship between two quantitative variables (e.g. job experience vs. salary)
= Streu/ Punktdiagramm
What lays on the X-axis of a scatterplot?
Explanatory / indipendente variable
What lays on the Y-axis of a scatterplot?
Response / dependent variable
What is the strength of a scatterplot?
How closely the points follow a clear form
Which 3 forms can be shown in a scatterplot?
- overall pattern
- Striking deviations
- linear relationship: roughly straight line
Which directions can the association in the scatterplot have?
Positive association: increase in one variable means increase in the other
Negative association: increase in one variable means decrease in the other
What does the correlation r (Wechselwirkung) measure?
direction and strength of the linear relationship between two quantitative variables
What do you know about the properties (Eigenschaften) of a correlation?
In which range do they lay?
How do they deal with outliers?
Which values are they using?
Always between -1 and 1
-1 strongest negative correlation
0 no relation
1 strongest positive correlation
not resistant to outliers
used standardized values; no change in r, when changing units of measurement
Name 6 kinds of correlation r
strong positive correlation r=1
moderate positive correlation 01
no correlation r=0
moderate negative correlation 0>r>-1
strong negative correlation r=-1
curvilinear relationship
What to use to analyze the relation of two quantitative variables?
Scatterplot
What to use to analyze the relation of a quantitative and a categorical variable?
5 options
Boxplot, stem-leaf plot, pie chart, bar chart or histogram
What to use to analyze the relation of two categorical variables?
Two-way-table
What are the characteristics of a two-way table?
What does it organize?
What does it show?
Organize two categorical variables:
- row variable: across ->
- column variable: down
The table shows counts (= absolute frequencies) or percent (=relative frequencies)
Which 3 types of distribution can a two-way-table have?
Joint distribution:
- distribution for each cell
- divide number in cell by total sample size
Marginal distribution:
- distribution for a single variable
- divide total of a variable by total sample size
Conditional distribution:
- relative frequencies
- eg. given that the person is a boy what is the probability his eyes are no green?
What happens if you have no association in a two-way table?
- knowledge about x does not give any information about probability of y
- Joint distribution can be calculated by multiplying marginal distribution and dividing the product by the table total
- relative frequency (Prozente) equal in all columns (automatically in all rows as well)
What happens if you have a perfect association?
- By knowing x you can perfectly predict y
-fields have to be 0 for perfect association to occur
relative frequencies 0% and 100% in one condition, and 100% and 0% in the other
What happens if you have an association?
relative frequency distribution different between the columns ( also goes for the rows)
Explain the term reliability in the case of interrupter reliability
Reliability: to what extend a repeated measurement yields the same score
Explain the special case of interrater reliability with an example
- 2 raters judge a same set of objects independently
- They agree or disagree on different objects and you want to find out how much they agree (=interrater reliability =agreement Kappa)
When do you have agreement Kappa k?
If the raters judged the objects randomly, there is still chance of agreement = expected agreement.
Kappa takes out the random factor via subtracting this expected random chance
π = (π΄πππ πππ£ππ β π΄ππ₯ππππ‘ππ ): (π β π΄ππ₯ππππ‘ππ )
What would happen to k if the teacher would agree in every case?
A-Observed would equal N making the nominator and denominator the same. Kappa would come out exactly as 1.
How is the measure of agreement if k<0.4
bad
How is the measure of agreement if 0,4
mediocre
How is the measure of agreement if k>_0,7
good
