Samenvatting theorie Kwantitatieve data-anlyse SPSS

SAMENVATTING SPSS KWANTI - RAPPORTAGE


Inhoud
1. Inhoud............................................................................................................................. 2
2. Waarom statistieken?..................................................................................................... 3
3. Applied statistical modelling in criminology....................................................................3
3.1 History of applied statistical modelling......................................................................3
4. Basisconcepten (herhaling)............................................................................................. 4
5. Inleiding tot SPSS............................................................................................................ 5
5.1 Het programma openen............................................................................................. 5
5.2 Korte rondleiding in SPSS..........................................................................................5
5.3 Data invoeren............................................................................................................ 5
5.4 Een nieuwe variabele aanmaken in SPSS..................................................................6
5.5 Variabelen hercoderen in SPSS..................................................................................6
5.6 Schalen aanmaken in SPSS........................................................................................7
5.7 Cases selecteren in SPSS........................................................................................... 7
5.8 De ‘Split File’ procedure.............................................................................................7
Deel 1: Beschrijvende statistiek.......................................................................................... 7
1. Beschrijvende statistiek..................................................................................................7
1.1 Meetniveaus.............................................................................................................. 7
1.2 Tabellen en grafeken................................................................................................8
1.2.1 Een frequentietabel opvragen in SPSS.................................................................8
1.2.2 Grafeken maken in SPSS.....................................................................................8
1.3 geen idee?................................................................................................................. 8
1.3.1 Variabelen numeriek beschrijven........................................................................8
1.3.2 Maten van centraliteit en spreiding in SPSS.........................................................9
1.3.3 Boxplot................................................................................................................ 9
1.4 Vormen van scoreverdelingen...................................................................................9
1.4.1 Herhaling theorie................................................................................................. 9
1.4.2 Maten van vorm in SPSS......................................................................................9
1.5 Standaarddeviatie................................................................................................... 10
1.6 Verbanden tussen variabelen: grafeken en tabellen...............................................11
1.7 Correlatiecoëfciënt................................................................................................. 12
1.7.1 Herhaling theorie............................................................................................... 12
1.7.2 Correlatiecoëfciënten in SPSS..........................................................................12
Deel 2: Inferentiële statistiek – signifcantietoetsen..........................................................13
1. Inleiding tot de inferentiële statistiek............................................................................13
1.1 Steekproef en populatie........................................................................................... 13
1.2 Steekproevenverdeling............................................................................................ 13
1.3 Betrouwbaarheidsintervallen en statistische signifcantie.......................................15

,2. Statistische signifcantie van de correlatiecoëfciënt...................................................16
2.1 Het verwerpen of aanvaarden van de nulhypothese...............................................16
2.2 Statistische signifcantie van de correlatiecoëfciënt in SPSS..................................16
3. T-toets.......................................................................................................................... 16
3.1 De afhankelijke t-toets............................................................................................. 16
3.1.1 Herhaling theorie............................................................................................... 16
3.1.2 De afhankelijke t-toets in SPSS..........................................................................17
3.2 De onafhankelijke t-toets......................................................................................... 17
3.2.1 Herhaling theorie............................................................................................... 17
3.2.2 De onafhankelijke t-toets in SPSS......................................................................18
4. De chikwadraattoets..................................................................................................... 19
4.1 De chikwadraattoets voor twee steekproeven in SPSS............................................21
4.2 Het partitioneren van kruistabellen.........................................................................21
5. De F-toets..................................................................................................................... 21
5.1 Herhaling theorie..................................................................................................... 21
5.2 Varianties opvragen in SPSS....................................................................................22
6. Variantieanalyse (ANOVA)............................................................................................ 23
6.1 De ongecorreleerde één-factor variantieanalyse in SPSS........................................23
6.2 Variantieanalyse voor afhankelijke waarnemingen of herhaalde metingen.............24
Deel 3: correlatiemethoden – factoranalyse en regressie.................................................24
1. Factoranalyse............................................................................................................... 25
1.1 Introductie factoranalyse......................................................................................... 25
1.2 Factoranalyse in SPSS.............................................................................................. 26
1.3 Factoranalyse en multidimensionaliteit...................................................................28
1.4 Cronbach’s alpha..................................................................................................... 28
1.5 Schalen aanmaken.................................................................................................. 29
2. Regressieanalyse.......................................................................................................... 29
2.1 Herhaling correlatie................................................................................................. 29
2.2 Enkelvoudige lineaire regressie...............................................................................30
2.3 Assumpties regressie............................................................................................... 34
2.4 Regressie in SPSS.................................................................................................... 34
2.5 Meervoudige lineaire regressie................................................................................35
EXAMEN............................................................................................................................ 37


1. Inhoud
- Toepassen van kwantitatieve analysetechnieken: uitvoeren van statistische
analyses en datatransformaties via menu-interface;
- Verantwoord kiezen tussen univariate beschrijvende parameters;
- Verantwoord kiezen tussen bivariate parameters: van kruistabellen tot de
regressie, correlatieanalyse en variantie-analyse;

, - Verkennen van de grenzen van bivariate statistiek: het meervoudige lineaire
regressiemodel, de logistische regressie-analyse, beginselen van de factor-
analyse;
- Aanmaken van schalen en evalueren van de schaalbetrouwbaarheid
Eindcompetenties:
- Resultaten van statistische analyses begrijpen, interpreteren en kritisch
evalueren;
- Verantwoord kiezen uit verschillende statistische technieken op basis van het
meetniveau van variabelen;
- Een statistisch model zelfstandig analyseren in SPSS;
- Zelfstandig transformeren en hercoderen van variabelen in SPSS;
- Vertalen van onderzoeksvragen in een statistisch model;
- Interpreteren van statistische signifcantieniveaus en betrouwbaarheidsintervallen.


2. Waarom statistieken?
- Describing data: how many banks get robbed every year?
- Making inferences starting from your data:
o Youth sample: 5% serious ofenders: how would that be in the population of
adolescents?
- Testing hypotheses in the social world: does poverty lead to crime?
- Waarom niet leuk:
o Matematische achtergrond, formules, berkeningen, examples of felds
other then you personal feld of interest




3. Applied statistical modelling in criminology
o Crime & Criminal Justice Statistics: describing crime / victimization /
ofending patterns and series in countries, regions, cities or
neighbourhoods
o Exploratory Analysis: exploring patterns that lead to the formulation of
hypotheses
o Explanatory Analysis: explaining diferences or changes using variables
derived from theories
o Primary data vs. secondary data:
 Primary: the data are collected by the researcher and the aim is to
conduct scientifc research
 Secondary: the data are collected by a scientist for scientifc
purpose, but the scientist can make use of them
o E.g.: data form administration on poverty: can be used to conduct
research, but are especially gathered for knowing how many people live on
welfare

3.1History of applied statistical modelling
- In the good old days
o You punched cards to run your statistics
o Sat at a terminal for a main frame and wrote syntax that wouldn’t work if
you left of a comma
o Went to dinner while your PC churned away at the data
o You used text editors to enter data
 That has all changed now
- Why use SPSS instead of other good programs: SAS, Stata, Mplus, R,..

, o Diferent versions of SPSS
 Spss has undergone quite some improvement
 Which version?
 Some problems were reported in old versions
 Signifcant improvement has been made since version 13
(graphs, complex models)
 SPSS has bevome IBM-SPSS
 In the book we use SPSS 19
 In this course we use SPSS 25
- What is statistical software?
o Powerfull tool allowing you to
 Create and use data bases
 Conduct simple analyses such as mean, mode and median
calculation
 Conduct more complex analyses such as regression models, factor
analysis and other ‘tools of the trade”
 Visualize the distribution (e.eg. bar chart, box plot,..) of variables
and relations between variables (e.g. scatterplot)
 Perform hypothesis testing (making inferences)
o Is not:
 Magic, because a program only performs what you have asked it to
do
 Intelligent: because you have to make the substantial interpretation
yourself.
 Cheap: prices range from 200-16000 USD (depending on options
chosen)


4. Basisconcepten (herhaling)
- Statistical units (cases): are the units we want to say something about
o These units can be of different aggregation level (individuals in surveys,
neugbourhood crime rates,…)
- Variables: characteristics of statistical units that can take several values (e.g.
repeated victimization: 3 times, 4 times,..)
- Variables and units are what statistical analysis is all about
- Dependent variables vs. independent variables
o Dependent: outcome variable  eg how many times did you steal
something from a shop;
o The score on the outcome variable is dependent on the score on (an)
independent variable(s) – e.g.: “How many times have you beaten op
someone – I act before I think
o Cf. hypothesis derived from self-control theory: low self-control  violent
ofending
- Asymmetric vs. symmetric relations
o Asymmetric: causal analysis
o Symmetric: covariance, no causality
o Some techniques are used to study asymeetric relationships, others to
study symmetric relationships
o You ALWAYS need to choos the appropriate technique: i.e. the best
technique of analysis that’ fts the research problem best