Inhoudsopgave
Unit 19 - Sampling ...................................................................................................................................... 2
Sampling .............................................................................................................................................................. 2
Sample and population ....................................................................................................................................... 4
Non probability sampling ................................................................................................................................... 8
Unit 20 – First steps towards inference: certainty about means .................................................................... 10
The sampling distribution ................................................................................................................................. 10
The central limit theorem ................................................................................................................................. 10
What is inferential statistics ............................................................................................................................. 11
Sampling distribution version 3 ........................................................................................................................ 12
Confidence intervals .......................................................................................................................................... 16
Paper; Analyzing data using linear models ...................................................................................................... 19
Q&A – Unit 19 & 20 ........................................................................................................................................... 21
Unit 19 - Sampling .................................................................................................................................... 23
Unit 20: certainty about the means ........................................................................................................... 25
Unit 21- First steps towards inference: effects and significance .................................................................. 31
First steps towards inference: certainty about means / the t-distribution..................................................... 31
First steps towards inference: certainty about slopes – Inference for slopes ................................................. 33
The regression line ............................................................................................................................................ 35
The regression equation ................................................................................................................................... 38
The regression model ........................................................................................................................................ 40
Analysing data using linear models - source.................................................................................................... 45
Unit 22 – Research ethics: ethics in social science research......................................................................... 46
Q&A 4/11/2021: Unit 21, 22, practise test 2 .............................................................................................. 48
>> Elaboration models: unit 14, 16, 17 ....................................................................................................... 48
1
,Unit 19 - Sampling
Sampling
When sampling?
- If not all units mentioned in our research question can be studied, we need to
‘sample’
- Studying a smaller set of units with the aim to say something about all units
1. Population = ‘Dutch people between 18 and 65 in 2015’
2. Sampling frame = population registry
→ form which we can select individuals
3. Sample = ‘every 100th individual’
→ selection process
4. Studied units = people participating
→ everyone who wants to participate
5. Data on studied units = people answering a specific question
→ not everyone is willing to answer every question
What is sampling?
Sampling process
=
1. Population
2. Sampling frame
3. Sample
4. Interviewed sample
5. Data
→ the step between step 2 and step 3 = sampling!
Distortions in sampling process
- Relationship between step 1 and step 2 → distortions by registration errors
- Relationship between step 2 and step 3 → distortions by sampling error & bias
- Relationship between step 3 and step 4 → distortions by non-response and refusals
- Relationship between step 4 and step 5 → distortions by item non-response
→ response rate = interviewed sample size / sample size → the higher the better
(= relationship between step 3 and step 4)
Sampling procedures
2
, Is the chance that a specific unit from the sampling frame is included in the study, known?
So, do we know the chance that an individual is included in our sample?
o No = non-probability
o Yes = probability sampling
Non-probability sampling
o Convenience sampling
= interview a few people you happen to know
o Purposive (judgmental) sampling
= units to be observed are selected on the basis of the researcher’s judgement about
which one’s will be the most useful or representative
o Snowball sampling
= interviewed person may be asked to suggest additional people for interviewing
o Quota sampling
= units are selected into a sample on the basis of prespecified characteristics, so that
the total sample will have the same distribution of characteristics assumed to exist in
the population being studied
Example of non-probability sampling: opt-in survey of some newspaper.
- Only those that read the newspaper fill out the questionnaire
- Selected units do NOT necessarily reflect the population. The sample is probably
‘biased’
- Not everyone has an equal chance to be included in the sample
Probability sampling
o Simple random sampling
o Stratified sampling
o (multi-stage) cluster sampling
Example of simple random sample from the population registry
(e.g. every 100th individual or use a random number)
- Selected units reflect the population
- Everyone has a known chance of being included in the sample:
if every 100th individual the chance that you get included is 1%
Assessing sampling
We always make sampling mistakes
Two types of mistakes:
- Sampling bias (sampling invalidity)
- Sampling error (sampling unreliability)
Sampling bias
- Bias = not being typical for the population. Studying the wrong group of people
Example snowball sampling:
‘How many people in the Netherlands currently support the EU?’
3
Unit 19 - Sampling ...................................................................................................................................... 2
Sampling .............................................................................................................................................................. 2
Sample and population ....................................................................................................................................... 4
Non probability sampling ................................................................................................................................... 8
Unit 20 – First steps towards inference: certainty about means .................................................................... 10
The sampling distribution ................................................................................................................................. 10
The central limit theorem ................................................................................................................................. 10
What is inferential statistics ............................................................................................................................. 11
Sampling distribution version 3 ........................................................................................................................ 12
Confidence intervals .......................................................................................................................................... 16
Paper; Analyzing data using linear models ...................................................................................................... 19
Q&A – Unit 19 & 20 ........................................................................................................................................... 21
Unit 19 - Sampling .................................................................................................................................... 23
Unit 20: certainty about the means ........................................................................................................... 25
Unit 21- First steps towards inference: effects and significance .................................................................. 31
First steps towards inference: certainty about means / the t-distribution..................................................... 31
First steps towards inference: certainty about slopes – Inference for slopes ................................................. 33
The regression line ............................................................................................................................................ 35
The regression equation ................................................................................................................................... 38
The regression model ........................................................................................................................................ 40
Analysing data using linear models - source.................................................................................................... 45
Unit 22 – Research ethics: ethics in social science research......................................................................... 46
Q&A 4/11/2021: Unit 21, 22, practise test 2 .............................................................................................. 48
>> Elaboration models: unit 14, 16, 17 ....................................................................................................... 48
1
,Unit 19 - Sampling
Sampling
When sampling?
- If not all units mentioned in our research question can be studied, we need to
‘sample’
- Studying a smaller set of units with the aim to say something about all units
1. Population = ‘Dutch people between 18 and 65 in 2015’
2. Sampling frame = population registry
→ form which we can select individuals
3. Sample = ‘every 100th individual’
→ selection process
4. Studied units = people participating
→ everyone who wants to participate
5. Data on studied units = people answering a specific question
→ not everyone is willing to answer every question
What is sampling?
Sampling process
=
1. Population
2. Sampling frame
3. Sample
4. Interviewed sample
5. Data
→ the step between step 2 and step 3 = sampling!
Distortions in sampling process
- Relationship between step 1 and step 2 → distortions by registration errors
- Relationship between step 2 and step 3 → distortions by sampling error & bias
- Relationship between step 3 and step 4 → distortions by non-response and refusals
- Relationship between step 4 and step 5 → distortions by item non-response
→ response rate = interviewed sample size / sample size → the higher the better
(= relationship between step 3 and step 4)
Sampling procedures
2
, Is the chance that a specific unit from the sampling frame is included in the study, known?
So, do we know the chance that an individual is included in our sample?
o No = non-probability
o Yes = probability sampling
Non-probability sampling
o Convenience sampling
= interview a few people you happen to know
o Purposive (judgmental) sampling
= units to be observed are selected on the basis of the researcher’s judgement about
which one’s will be the most useful or representative
o Snowball sampling
= interviewed person may be asked to suggest additional people for interviewing
o Quota sampling
= units are selected into a sample on the basis of prespecified characteristics, so that
the total sample will have the same distribution of characteristics assumed to exist in
the population being studied
Example of non-probability sampling: opt-in survey of some newspaper.
- Only those that read the newspaper fill out the questionnaire
- Selected units do NOT necessarily reflect the population. The sample is probably
‘biased’
- Not everyone has an equal chance to be included in the sample
Probability sampling
o Simple random sampling
o Stratified sampling
o (multi-stage) cluster sampling
Example of simple random sample from the population registry
(e.g. every 100th individual or use a random number)
- Selected units reflect the population
- Everyone has a known chance of being included in the sample:
if every 100th individual the chance that you get included is 1%
Assessing sampling
We always make sampling mistakes
Two types of mistakes:
- Sampling bias (sampling invalidity)
- Sampling error (sampling unreliability)
Sampling bias
- Bias = not being typical for the population. Studying the wrong group of people
Example snowball sampling:
‘How many people in the Netherlands currently support the EU?’
3
