Definitions

Question 1

Test statistic

Answer 1

Compares how good the model is (variance explained by model) with how bad it is (variance not explained by the model)
–> signal to noise ratio

—> p-wert errechnung, auf F

Question 2

independent variable

Answer 2

the proposed cause, the predictor, manipulated variable

Question 3

dependent variable

Answer 3

the proposed effect, the measured outcome variable

Question 4

Whats a categorial variable?

Answer 4

divided into distinct categories
*binary: two options
*nominal: 3+ options
*ordinal: options with logical order

Question 5

Whats a continuous variable?

Answer 5

entities get a distinct score
*interval: equal intervals in variable (no true zero e. g. temperature)
*ratios of scores on scale must make sense (true zero e. g. weight)

Question 6

standard deviation

Answer 6

measures how spread out /dispersed the data are around the mean
–> standard deviation is the square root of the variance

Question 7

statistical model

Answer 7

a representation of a real world process designed to predict outcomes and understand relationships between variables

Question 8

linearity

Answer 8

Linearity means that two variables have a straight-line relationship, where changes in one variable correspond proportionally to changes in the other

Question 9

additivity

Answer 9

Additivity assumes that the effects of predictors on the outcome combine linearly, with no interactions, so their total effect is the sum of individual effects.

Question 10

central limit theorem

Answer 10

if the sample is big enough (above 30 data points), the parameter estimations will have a normal distribution

Question 11

parameters

Answer 11

explain the relationsship between outcome and predictor
–> “b”

Question 12

p-value

Answer 12

Shows the likelihood of the result if the null hypothesis is true; p < 0.05 means statistically significant.

Question 13

r (correlation coefficient)

Answer 13

Measures the strength and direction of a linear relationship; ranges from -1 (negative) to +1 (positive).

Question 14

R² (coefficient of determination)

Answer 14

Indicates the proportion of variance explained by the model; higher values mean better prediction.

Question 15

Confidence interval (CI)

Answer 15

A range where the true parameter likely falls; a 95% CI includes the true value 95% of the time.

Question 16

Chi-square statistic

Answer 16

Tests independence or fit in categorical data; larger values show greater deviation from expectations.

Question 17

F-statistic

Answer 17

f represents the overall significance of the model (comparing variance explained by the model with unexplained variance)

high f means Strong Model Fit

Question 18

odds ratio

Answer 18

measure of effect size for categorical data

Die Odds Ratio gibt an, wie viel wahrscheinlicher ein Ereignis in einer Gruppe im Vergleich zur Referenzgruppe ist.
(OR = 1: kein Unterschied; OR > 1: höhere Wahrscheinlichkeit; OR < 1: niedrigere Wahrscheinlichkeit).

Für chi :

OR = 1: Kein Unterschied zwischen den Gruppen.
OR > 1: Höhere Wahrscheinlichkeit in der untersuchten Gruppe.
OR < 1: Niedrigere Wahrscheinlichkeit in der untersuchten Gruppe.

Question 19

standardized residulas

Answer 19

shows how far from the reality the model is (wenn insignifikant also werte zwischen +- 1,96 dann guter model fit)

Question 20

normality

Answer 20

the sampling distribution of the parameter estimate should be normally distributed

-> symmetrisch und glockenförmig um ihren Mittelwert

Question 21

z-score

Answer 21

Measures how many standard deviations a value is from the mean.

Vergleichbarkeit herstellen
Ausreißer erkennen
Normalverteilung prüfen

z = (Wert - Mittelwert) / Standardabweichung