CMR MASTER DOCUMENT for final exam 9:30 – 11:00 Tuesday 24th THEORY
LECTURE 2
QUANTITATIVE RESEARCH QUALITATIVE RESEARCH
• Exploration & description
• Insights on complex, new, and sensitive phenomena
• Exploration & explanation • General open-ended questions
• Test specific hypotheses • Small samples of informants
• Usually close(d) questions • Aims to interpret
• Large(r) samples of informants • Theoretical framework is evolving & emerging from data
• Aims to generalize (statistical inference)
• Theoretical framework is tested on data Why qualitative research? Preferences and/or experience of the
researcher, Preferences and/or experience of the research user,
sensitive information, complex phenomen, developing new theory
QUALITATIVE research techniques:
Primary data
Secondary data
General QUALITATIVE data analysis procedure:
1. Data collection – gather data from a variety of sources. Transcribe / record everything said and who says
what. Transcribe things not said, e.g. gestures, attitudes, etc. The richer the data, the more insightful the
analysis can be.
2. Data Analysis – In deductive coding approaches, researchers develop a codebook and code the data (e.g., find
the excerpts that fit the codes). Then, researchers organizing codes into higher-level categories or themes,
structuring in a framework.
3. Data verification – check for alternative explanations. challenge is subjectivity.
dplyr - make transformations
Creating new datasets based on an existing dataset 1. Filtering out participants 2. Selecting variables
Within a dataset: 1. Creating new variables as functions of old variables 2. Recoding old variables into new variables
Filtering out participants:
Creating new variables as functions of old variables
Recoding old variables into new variables
, LECTURE 3
Sampling error : Statistics of the sample (e.g., mean age) generally differ from the entire population.
Sampling errors corresponds to the difference between the sample statistic and population parameter.
Sampling error is always present when making statements about a population
To reduce sampling error: increasing sample size. The larger the sample, the lower the sampling error
Sampling bias: systematic tendency’ to sampling error
• Self-selection bias: participants with specific characteristics are more likely to take part in a study than others.
• Pre-screening: the way participants are pre- screened may lead to a sample bias
• Under-coverage: specific members of the population are inadequately represented in the sample
• Survivorship: successful observations are more likely to be represented in the sample than unsuccessful ones
Sampling bias is a threat to external validity, it limits to generalizability of research findings to the broader population.
stratified sampling: Divided into sub- groups based on distinguishing factors
(strata), then random sampling in each one
LECTURE 2
QUANTITATIVE RESEARCH QUALITATIVE RESEARCH
• Exploration & description
• Insights on complex, new, and sensitive phenomena
• Exploration & explanation • General open-ended questions
• Test specific hypotheses • Small samples of informants
• Usually close(d) questions • Aims to interpret
• Large(r) samples of informants • Theoretical framework is evolving & emerging from data
• Aims to generalize (statistical inference)
• Theoretical framework is tested on data Why qualitative research? Preferences and/or experience of the
researcher, Preferences and/or experience of the research user,
sensitive information, complex phenomen, developing new theory
QUALITATIVE research techniques:
Primary data
Secondary data
General QUALITATIVE data analysis procedure:
1. Data collection – gather data from a variety of sources. Transcribe / record everything said and who says
what. Transcribe things not said, e.g. gestures, attitudes, etc. The richer the data, the more insightful the
analysis can be.
2. Data Analysis – In deductive coding approaches, researchers develop a codebook and code the data (e.g., find
the excerpts that fit the codes). Then, researchers organizing codes into higher-level categories or themes,
structuring in a framework.
3. Data verification – check for alternative explanations. challenge is subjectivity.
dplyr - make transformations
Creating new datasets based on an existing dataset 1. Filtering out participants 2. Selecting variables
Within a dataset: 1. Creating new variables as functions of old variables 2. Recoding old variables into new variables
Filtering out participants:
Creating new variables as functions of old variables
Recoding old variables into new variables
, LECTURE 3
Sampling error : Statistics of the sample (e.g., mean age) generally differ from the entire population.
Sampling errors corresponds to the difference between the sample statistic and population parameter.
Sampling error is always present when making statements about a population
To reduce sampling error: increasing sample size. The larger the sample, the lower the sampling error
Sampling bias: systematic tendency’ to sampling error
• Self-selection bias: participants with specific characteristics are more likely to take part in a study than others.
• Pre-screening: the way participants are pre- screened may lead to a sample bias
• Under-coverage: specific members of the population are inadequately represented in the sample
• Survivorship: successful observations are more likely to be represented in the sample than unsuccessful ones
Sampling bias is a threat to external validity, it limits to generalizability of research findings to the broader population.
stratified sampling: Divided into sub- groups based on distinguishing factors
(strata), then random sampling in each one
