TENTAMEN: College Notes and Summary - Advanced Research Methods and Statistics ()

ARMS – Samenvatting
Inhoud
TOE – Hoorcollege 4 – 16-2-2023...........................................................................................................2
Refreshing Correlation and Regression..................................................................................................5
Refreshing T-test and ANOVA.................................................................................................................7
Grasple – Refresh Lessons – Part 1........................................................................................................9
Grasple – Refresh Lessons – Part 2......................................................................................................10
Grasple – Refresh Lessons – Part 3......................................................................................................11
Grasple – Week 1.................................................................................................................................13
Hoorcollege 1 – 13-11-2023.................................................................................................................16
Werkgroep 1 – 16-11-2023..................................................................................................................18
Hoorcollege 2 – 16-11-2023.................................................................................................................20
Grasple – Week 2.................................................................................................................................22
Hoorcollege 3 – 20-11-2023.................................................................................................................26
Experimentwise Error Rate..................................................................................................................27
Werkgroep 2 – 23-11-2023..................................................................................................................28
Hoorcollege 4 – 23-11-2023.................................................................................................................29
Grasple – Week 3.................................................................................................................................30
Hoorcollege 5 – 27-11-2023.................................................................................................................34
Werkgroep 3 – 30-11-2023..................................................................................................................36
Hoorcollege 6 – 30-11-2023.................................................................................................................39
Grasple – Week 4.................................................................................................................................41
Hoorcollege 7 – 4-12-2023...................................................................................................................45
Werkgroep 4 – 7-12-2023....................................................................................................................48
Hoorcollege 8 – 7-12-2023...................................................................................................................49
Grasple – Week 5.................................................................................................................................54
Hoorcollege 9 – 11-12-2023.................................................................................................................55
Werkgroep 5 – 14-12-2023..................................................................................................................58




1

,TOE – Hoorcollege 4 – 16-2-2023
Measurement and Bias in Regression:

- Survey Method:
o Issues with reliability due to recall and estimation.
o Limited construct validity.
- Lab Experiment:
o Problems with altered behavior (Hawthorne effect).
o Difficulty in maintaining 24-hour observation.
- Pictures:
o Concerns about construct validity and reliability.
o Offer contextual information.
- Diary:
o Challenges with non-response, recall, and under-reporting.
o Traditional and modern diary methods.
- Measurement Issues:
o Difficulty in measuring accurately.
o Importance of validity and reliability.




- Response Process Challenges:
- Survey Bias:
o Recall problems leading to bias.
o Bias sources in the survey lifecycle.
- Total Survey Error Framework:
o Components:
 Measurement error: the inaccuracy or imprecision in the process of
collecting data or measurements (errors in data entry, coding, handling
confidential information).
 Processing error: mistakes, inaccuracies, or oversights that occur during the
handling, coding, entry, or analysis of data after it has been collected.
 Coverage error: when not all members of the population have a known, non-
zero chance of being included in the sample.









2

, 
 Sampling error: when a survey samples only some, rather than all, members
of the covered population.
 Non-response error: when selected individuals refuse to cooperate with the
survey (unit non-response) or decline to answer specific questions (item
non-response).





 Adjustment error: when adjustments are made to correct for non-response
or coverage issues in a survey, but these adjustments introduce bias or
inaccuracies in the final survey estimates (required to correct
underrepresentation).
- Question bias (survey-related) and response bias (respondent behavior).
o Response Biases:
 Social desirability: respondents may provide answers that they perceive as
socially acceptable or favourable rather than expressing their true beliefs or
behaviours.
 Acquiescence: the tendency of respondents to agree with statements or
questions regardless of their actual beliefs, possibly to please the surveyor or
avoid conflict.
 Primacy/recency effects: the order in which questions are presented may
influence respondents' answers. Primacy effect occurs when earlier items
have more impact, while recency effect occurs when later items are more
influential.
 Straight-lining: respondents provide the same response (e.g., selecting the
same Likert scale option) consistently throughout a survey without
differentiation.
 Fence-sitting: respondents consistently choose neutral or middle response
options, avoiding extreme positions.
- Matrix questions and strategies to avoid biases.
o Use of answer categories for multiple questions.
o Comparison of matrix and individual questions.

Research cycle:


3

, 

Survey lifecycle









4