Inhoudsopgave
HC 1: Intro.............................................................................................................................................................2
HC 2: Statistische Modellen..................................................................................................................................5
HC 3: SPINE and Assumptions.............................................................................................................................10
PU3: Grafieken en assumpties testen...........................................................................................................19
HC 4: NHST, Assumpties en Data Visualisatie (HERHALING)..............................................................................21
PU4: Recoding and Computing variables / Reliability...................................................................................24
HC 5: Comparing means – independent/dependent/one sample t-test............................................................24
PU5: T-tests....................................................................................................................................................33
HC 6: Comparing means – One-way ANOVA, Reliability....................................................................................37
PU6: One-way ANOVA...................................................................................................................................43
HC 7: Comparing means – One-way ANOVA (Follow up analyses), Factorial ANOVA PART 1 & Interactie
effecten...............................................................................................................................................................47
PU7: Factorial ANOVA....................................................................................................................................51
HC 8: Comparing means – Factorial ANOVA PART 2..........................................................................................54
H9: Chi Square test.............................................................................................................................................56
PU9: Chi-Square Test.....................................................................................................................................62
HC 10: Correlatie.................................................................................................................................................64
PU10: Correlation coefficients.......................................................................................................................70
HC 11: Lineaire Regressie PART 1.......................................................................................................................71
PU11: Lineaire Regressie...............................................................................................................................76
HC 12: Lineaire Regressie: Assumpties PART 2...................................................................................................78
HC 13: RECAP: Sammenvatting van alle colleges...............................................................................................81
1
,HC 1: Intro
Falsificatie
Verwerpen van de theorie of hypothese
You can only examine whether a theory/hypothesis is credible, if there is a possibility
to disproof it.
The principle of hypothesis testing
Women are more intelligent than men.
N = 2, men score 108, women score 109.
Is my hypothesis supported or not? What if N = 10? 100? 1000? 10.000? 100.000?
Point of departure = assumption that there is no difference
This gives a point of comparison
If no difference, than IQ(women) – IQ(men) = 0
We can predetermine:
o If I measure IQ in 1000 persons, and the mean difference between men and
women is larger than 5 IQ-points, then it is very unlikely that this difference is
‘coincidence’.
Typen hypotheses
Nulhypothese, H0
o Er is geen effect
o Women are equally likely as men to wear a skirt or dress
o No relationship between age and the number of wrinkles
Alternatieve hypothese, H1
o Er is verschil gevonden
o Women are more likely to wear a skirt or dress than man
o There is a positive relationship between age and the number of wrinkles you
have: The older people are, the more wrinkles they have.
o If we can reject H0, this H1 is SUPPORTED by the data but not PROVEN
Why do we need statistics?
Statistics offer us a means to determine exactly how (un-)likely (on-waarschijnlijk) it
is that we would observe a set of data if the null hypothesis were true.
o If it is very unlikely (chances are smaller than 5%) we may conclude that there
is support for our alternative hypothesis.
In other words: We examine the probability the null hypothesis is true.
In this light, statistics is a form of argumentation
Variabelen
A variable ‘varies’
It has different values
o “Women who spend a lot of time following fitgirls are less satisfied with their
body than women who spend less time following fitgirls.”
Two important questions you need to be able to answer:
o What is the dependent variable, what is the independent variable?
2
, o What is the measurement level of my variables? (This is especially relevant for
the dependent variable)
Experiment
Je manipuleert iets
Dit moet een effect hebben
In andere woorden… OORZAAK EFFECT
Correlationeel design
Je meet/observeert waargenomen realiteit
o Kijken naar associaties
Bijv. is depressie geassocieerd met slechte gezondheid?
o Voorspelling outcome variable
Bijv. Voorspelt deelname colleges het cijfer?
Terminologie
Onafhankelijke Variable
Bij een experiment: de veronderstelde oorzaak, wordt gemanipuleerd
Bij een vragenlijst: a predictor variable: voorspeller
Afhankelijke Variable
Bij een experiment: het veronderstelde effect
Bij een vragenlijst: een outcome variable: uitkomst
Gemeten, niet gemanipuleerd
Measurement levels
Two main categories:
Categorical variables
o entities are divided into distinct categories
o E.g., gender
Continuous variables
o entities get a distinct score
o E.g., age
Important distinction because they determine which analysis you can do
3
, Levels of Measurement
Categorical (entities divided into distinct categories):
1. Binary or dichotomous variable: There are only two categories
a. e.g. dead or alive, pregnant or not pregnant
2. Nominal variable: There are more than two categories
a. e.g. omnivore, vegetarian or vegan, or fruitarian.
Binary and Nominal variables only allow you to say whether something equals something or
not (Plus, you can count how many there are):
7 = Cristiano Ronaldo
23 ≠ 7
Mezut Özil ≠ Cristiano Ronaldo
3. Ordinal variable: The same as a nominal variable but the categories are meaningfully
ordered
a. a fail, a pass, a merit or a distinction in their exam.
b. completely agree, agree, disagree or completely disagree
Ordinal variables allow you to say something about the order of things (order): 5th child
comes later than 1st or 4th child
They also allow you to say whether something equals something or not (equality): 5th
child ≠ 1st child
Continuous (entities get a distinct score):
1. Interval variable: Equal intervals on the variable represent equal differences in the
property being measured
Interval variables allow you to say something about the distance between units:
1940 – 1945 is a five-year period, just as 2005 – 2010.
They also allow you to say something about the order of things (order): 2010 comes later
than 2005
And they allow you to say something about equality: 2010 ≠ 1940
Interval variables do not have a ‘real’ 0 point. Year 0 does
not mean ‘the absence of time’.
2. Ratio variables: The same as an interval variable, but the ratios of scores on the scale
must also make sense
a. Weight in kilograms
b. Number of calories in food o Temperature (in Kelvin)
Waarom is dit belangrijk?
o Het meetniveau bepaalt wat je met een variabele kan doen (en daarmee de
statistische toetsen die je kan gebruiken)
o Categorisch: tellen
o Ordinaal: tellen, ordenen
o Interval: tellen, ordenen, optellen, aftrekken
o Ratio: tellen, ordenen, optellen, aftrekken, vermenigvuldigen, etc.
4
HC 1: Intro.............................................................................................................................................................2
HC 2: Statistische Modellen..................................................................................................................................5
HC 3: SPINE and Assumptions.............................................................................................................................10
PU3: Grafieken en assumpties testen...........................................................................................................19
HC 4: NHST, Assumpties en Data Visualisatie (HERHALING)..............................................................................21
PU4: Recoding and Computing variables / Reliability...................................................................................24
HC 5: Comparing means – independent/dependent/one sample t-test............................................................24
PU5: T-tests....................................................................................................................................................33
HC 6: Comparing means – One-way ANOVA, Reliability....................................................................................37
PU6: One-way ANOVA...................................................................................................................................43
HC 7: Comparing means – One-way ANOVA (Follow up analyses), Factorial ANOVA PART 1 & Interactie
effecten...............................................................................................................................................................47
PU7: Factorial ANOVA....................................................................................................................................51
HC 8: Comparing means – Factorial ANOVA PART 2..........................................................................................54
H9: Chi Square test.............................................................................................................................................56
PU9: Chi-Square Test.....................................................................................................................................62
HC 10: Correlatie.................................................................................................................................................64
PU10: Correlation coefficients.......................................................................................................................70
HC 11: Lineaire Regressie PART 1.......................................................................................................................71
PU11: Lineaire Regressie...............................................................................................................................76
HC 12: Lineaire Regressie: Assumpties PART 2...................................................................................................78
HC 13: RECAP: Sammenvatting van alle colleges...............................................................................................81
1
,HC 1: Intro
Falsificatie
Verwerpen van de theorie of hypothese
You can only examine whether a theory/hypothesis is credible, if there is a possibility
to disproof it.
The principle of hypothesis testing
Women are more intelligent than men.
N = 2, men score 108, women score 109.
Is my hypothesis supported or not? What if N = 10? 100? 1000? 10.000? 100.000?
Point of departure = assumption that there is no difference
This gives a point of comparison
If no difference, than IQ(women) – IQ(men) = 0
We can predetermine:
o If I measure IQ in 1000 persons, and the mean difference between men and
women is larger than 5 IQ-points, then it is very unlikely that this difference is
‘coincidence’.
Typen hypotheses
Nulhypothese, H0
o Er is geen effect
o Women are equally likely as men to wear a skirt or dress
o No relationship between age and the number of wrinkles
Alternatieve hypothese, H1
o Er is verschil gevonden
o Women are more likely to wear a skirt or dress than man
o There is a positive relationship between age and the number of wrinkles you
have: The older people are, the more wrinkles they have.
o If we can reject H0, this H1 is SUPPORTED by the data but not PROVEN
Why do we need statistics?
Statistics offer us a means to determine exactly how (un-)likely (on-waarschijnlijk) it
is that we would observe a set of data if the null hypothesis were true.
o If it is very unlikely (chances are smaller than 5%) we may conclude that there
is support for our alternative hypothesis.
In other words: We examine the probability the null hypothesis is true.
In this light, statistics is a form of argumentation
Variabelen
A variable ‘varies’
It has different values
o “Women who spend a lot of time following fitgirls are less satisfied with their
body than women who spend less time following fitgirls.”
Two important questions you need to be able to answer:
o What is the dependent variable, what is the independent variable?
2
, o What is the measurement level of my variables? (This is especially relevant for
the dependent variable)
Experiment
Je manipuleert iets
Dit moet een effect hebben
In andere woorden… OORZAAK EFFECT
Correlationeel design
Je meet/observeert waargenomen realiteit
o Kijken naar associaties
Bijv. is depressie geassocieerd met slechte gezondheid?
o Voorspelling outcome variable
Bijv. Voorspelt deelname colleges het cijfer?
Terminologie
Onafhankelijke Variable
Bij een experiment: de veronderstelde oorzaak, wordt gemanipuleerd
Bij een vragenlijst: a predictor variable: voorspeller
Afhankelijke Variable
Bij een experiment: het veronderstelde effect
Bij een vragenlijst: een outcome variable: uitkomst
Gemeten, niet gemanipuleerd
Measurement levels
Two main categories:
Categorical variables
o entities are divided into distinct categories
o E.g., gender
Continuous variables
o entities get a distinct score
o E.g., age
Important distinction because they determine which analysis you can do
3
, Levels of Measurement
Categorical (entities divided into distinct categories):
1. Binary or dichotomous variable: There are only two categories
a. e.g. dead or alive, pregnant or not pregnant
2. Nominal variable: There are more than two categories
a. e.g. omnivore, vegetarian or vegan, or fruitarian.
Binary and Nominal variables only allow you to say whether something equals something or
not (Plus, you can count how many there are):
7 = Cristiano Ronaldo
23 ≠ 7
Mezut Özil ≠ Cristiano Ronaldo
3. Ordinal variable: The same as a nominal variable but the categories are meaningfully
ordered
a. a fail, a pass, a merit or a distinction in their exam.
b. completely agree, agree, disagree or completely disagree
Ordinal variables allow you to say something about the order of things (order): 5th child
comes later than 1st or 4th child
They also allow you to say whether something equals something or not (equality): 5th
child ≠ 1st child
Continuous (entities get a distinct score):
1. Interval variable: Equal intervals on the variable represent equal differences in the
property being measured
Interval variables allow you to say something about the distance between units:
1940 – 1945 is a five-year period, just as 2005 – 2010.
They also allow you to say something about the order of things (order): 2010 comes later
than 2005
And they allow you to say something about equality: 2010 ≠ 1940
Interval variables do not have a ‘real’ 0 point. Year 0 does
not mean ‘the absence of time’.
2. Ratio variables: The same as an interval variable, but the ratios of scores on the scale
must also make sense
a. Weight in kilograms
b. Number of calories in food o Temperature (in Kelvin)
Waarom is dit belangrijk?
o Het meetniveau bepaalt wat je met een variabele kan doen (en daarmee de
statistische toetsen die je kan gebruiken)
o Categorisch: tellen
o Ordinaal: tellen, ordenen
o Interval: tellen, ordenen, optellen, aftrekken
o Ratio: tellen, ordenen, optellen, aftrekken, vermenigvuldigen, etc.
4
