TOE hoorcollege slides correlationeel

HC1: surveys
KOM overview
- Qualitative research
o Study people in natural environment
o Holistic approach
o Interviews, focus groups, text analysis
- Correlational research
o Quantitative
o Relationships between variables
o Causal study difficult
- Experimental research
o Researcher manipulation
o Experimental and control group
o Quantitative measures
o Suitable for causal research 3

TOE overview
- Qualitative research
o 5 lectures
o Nvivo Labs as homework
- Correlational research
o 5 lectures
o SPSS Labs as homework
- Experimental research
o 5 lectures
o JASP Labs as homework
- SPSS/JASP Exam
- Final Exam 5




Meer digitale data door digitaal tijdperk

,Correlational Data
Data are generated in different ways:

- Incidentally  niet met oog op onderzoek en er word ook niet gevraagd of je mee wil doen
met onderzoek (dingen posten op sociale media, maar ook pinpas)
o AKA organic
- Purposively  met het oog op onderzoek
o AKA designed




13

Correlational Data (designed)
We design a study & collect data to:

- describe the social reality we willen leren hoe het eraan te gaat in de maatschappij, zijn er
verschillen tussen verschillende regio’s in een land, zijn er dingen die we kunne beïnvloeden
- study (causal) relationships
- generalize to the target population  wat we leren uit de stekproef kunnen koppelen aan
de grote groep

Inferential goals

- Description
- Causation
- Prediction  kunnen we have data modellen bouwen om dingen te voorspellen

,How to ask? Survey Modes
- Face-to-face (CAPI)
o Kinderen die een vragenlijst moeten invullen
o Computer assisted personal interview
- Mail
o Post
o Geen email!
- Telephone (CATI)
o Computer assisted telephone interview
o Kirsten wordt altijd gebeld als ze aan het koken is
- Internet
o email
- Mixed-mode
o Verschillende vormen aan elkaar koppelen



Differences between the modes
- Degree of interviewer involvementhoeveel is de interviewer betrokken, is het belangrijk
dat de onderzoeker er bij is?
- Degree of interaction with the respondent is het belangrijk dat er interactie is
- Degree of privacy anoniem invullen?
- Channels of communication
o Visual
o Auditive
- Technology use

Survey modes in comparison
attribute Face-to-face mail telephone Internet
Cost high low moderate Low
Response rate High low high Moderate
Researcher high Low moderate moderate
control over
interview
Interviewer high low Moderate low
effect

, Survey Modes in NL
- Population register & over 90% internet use
o Telephone Surveys (Random Digit Dialing amerika heeft geen gemeente register)
not widely used
- Mixed modes
o Postal mail with invitations to internet surveys
o Telephone component if numbers are known
- Market Researchers use both probability and nonprobability online panels
- Mixed-device surveys
o Computer / smartphone

Types of mixed-mode design
One mode for some respondents, another mode for others E.g. online survey with mail
component for those without Internet

One mode for recruitment, another for survey administration E.g. mail invitation for an online
survey

One mode for data collection, another for reminders, follow-up-E.g. telephone reminders for an
online survey

One mode for main part of the interview, another for subset of answers (e.g. sensitive)E.g.
Telephone & Audio Computer Self-Administered (ACASI)

Cross-sectional and panel surveys
- Panel surveys interview respondents over time
o content usually the same
o but may differ
- Advantages:
o We can assess within-person change & causality
o We can disentangle age, period, and cohort effects  andere groep in andere
setting (over de Jaren heen)

Cross-sectional and panel surveys
Potential errors:

- Attrition
o drop-out
o consecutive wave-nonresponse
- Panel conditioning
o AKA learning effects ieder jaar dezelfde krijg je dezelfde antwoorden door dat
mensen ontouden wat ze antwoorden

Types of mixed-mode design
One mode for one wave of the panel survey, another for others E.g. First wave face-to-face,
following waves online to save costs

Operational process
- To measure a (theoretical) construct, researchers must follow steps: