Section C: Data Handling and Analysis Flashcards
i) What is ‘qualitative data’? - is it numerical?
ii) How can it be collected and analysed?
iii) What are some characteristics of qualitative data?
iiii) Where is it often collected?
i) Qualitative data is imprecise, non-numerical data (e.g. text, video, recordings).
ii) Qualitative data can be collected using diary accounts or in-depth interviews, and analysed using thematic analysis (grouped according to themes).
iii) Rich in detail, often subjective. Often used to reveal attitudes, beliefs and opinions. Low in reliability.
iiii) Often collected in real life settings.
i) What is ‘quantitative data’ - numerical?
ii) How is quantitive data analysed?
iii) What are some characteristics of quantitative data?
iiii) Where is it often collected?
i) Quantitative data is PRECISE, NUMERICAL data that can be quantified. It can be counted or measured.
ii) Quantitative data lends itself to statistical analysis, while qualitative data is grouped according to themes.
iii) Objective. Brief and concise, limited –> low in detail. Not used to reveal attitudes, beliefs and opinions. High in reliability.
iiii) Collected in artificial, controlled settings.
Would data collection through observations via the use of behaviour categories and ratings of behaviour produce qualitative or quantitative data?
QUANTITATIVE (tallies).
A recovering patient describes his experience of schizophrenia - would this produce qualitative or quantitative data?
QUALITATIVE (rich in detail, subjective).
Do case studies usually produce qualitative or quantitative data?
Qualitative
(A03) Strength of QUANTITATIVE DATA. (think is it scientifically objective?)
P: Quantitative data is scientifically objective.
E: Quantitative data is more objectively analysed than qualitative data. This is because numerical data can be interpreted using statistical analysis.
E: This is a strength as there is less interpretation involved using quantitative data and comparisons can be drawn more easily between data sets. This is because statistical analysis are based on the principles of mathematics and allow researchers to objectively conclude whether statistically significant relationships or differences have been found.
L: This means analysis is free from bias and interpretation, so high in objectivity.
How does qualitative data compare in terms of being scientifically objective?
P: Qualitative data is highly SUBJECTIVE.
E: This is because qualitative data is non-numerical and includes data that is rich in detail and therefore subjective.
E: As such, data cannot be easily compared or categorised
L: This means analysis is open to bias and interpretation, so is scientifically subjective.
(A03) Strength of QUANTITATIVE Data (REPLICABILITY- how much researcher involvement?)
P: QUANTITATIVE data can be very easily replicated
E: This is because it is based on measured, NUMERICAL values.
E: Such data requires minimal involvement from researchers.
L: So a consistent analysis by multiple researchers is highly reliable.
(A03) Strength of QUALITATIVE Data (depth of detail/insight).
P: QUALITATIVE data is highly valid.
E: This is because it is based on non-numeric, detailed responses / accounts.
E: This is a strength as such data is in-depth and insightful.
L: Therefore, a researcher has the opportunity to capture rich, descriptive data about how people think and behave.
(A03) Limitation of QUALITATIVE Data - how does it compare in terms of depth of detail/insight?
P: QUANTITATIVE data is less valid.
E: This is because it is based on numeric data and data that is quantifiable.
E: This is a limitation as such data is narrow and lacks DEPTH of detail and the nature of turning thoughts and feelings into numbers can be seen as superficial.
L: Researchers therefore have less opportunity to capture rich data about how people think and behave.
Give an example of how QUALITATIVE Data lacks validity?
(E.g. closed questions limit your response to prescribed options).
When gathering quantitative data, respondents may be forced to select answers which do not represent their real thoughts and feelings, (YES, NO, IN THE MIDDLE) leading to data which is superficial, lacks detail and therefore has lower validity.
In what type of setting is qualitative data likely to be gathered?
QUALITATIVE data is likely to have been gathered in more natural environments (e.g. a researcher carrying out a case study of a person’s experience of mental illness would make use of a RANGE of qualitative methods such as interviews, observations in hospital/ at home).
How is this setting a STRENGTH of qualitative data? (setting qualitative data is likely to be gathered in).
P: Qualitative data is highly valid.
E: This is because QUALITATIVE data is likely to have been gathered in more natural environments (e.g a researcher carrying out a case study of a person’s experience of mental illness would make use of a range of qualitative methods such as interviews, observations in hospital / at home).
E: This is a strength because in natural environments, participants are more likely to behave more naturally/ more relaxed, meaning they are more likely to give honest/realistic answers and their behaviour is likely to be more representative of the real world.
L: Therefore, the findings are more likely to be credible and valid.
In what type of setting is QUANTITATIVE data likely to be gathered?
In artificial, controlled environments (i.e. a lab).
How is this a limitation of quantitative data? (setting quantitative data is collected in).
P: QUANTITATIVE data is less valid.
E: This is because QUANTITATIVE data is likely to have been gathered in artificial, controlled environments (E.g. researchers looking at the impact of rehearsal on memory might recall the NUMBER OF ITEMS RECALLED CORRECTLY from a list).
E: This is a weakness as if carried out in unnatural environments, ppts are likely to behave unnaturally as opposed to natural environments.
L: Therefore, findings may not be credible from quantitative data.
Which type of data gathering is more TIME and COST effective?
QUANTITATIVE DATA.
How is this a STRENGTH of quantitative data? What does this allow for? (think practically - cost/time effective?)
P: QUANTITATIVE data is more time and cost effective to gather.
E: This is because methods that gather QUANTITATIVE data immediately produce information that is numerical from large sample sizes (e.g. questionnaires that are largely distributed).
E: This is a strength as data can be easily compared and analysed without transformation.
L: Therefore, so such methods produce a large amount of data fairly quickly.
Whereas how does qualitative data compare? (in terms of practicality - is it cost and time efficient?)
P: Gathering QUALITATIVE data is less time and cost effective.
E: This is because methods that gather QUALITATIVE data produce information that has to be transformed before analysis can be carried out (i.e. thematic)
E: This is a weakness as transforming data into categories can be a lengthy and subjective process.
L: Therefore, this means that such methods may be more difficult to run.
What type of data is more useful in making broad generalisations?
Quantitative —> We are able to generate conclusions/generalisations from data. More nomothetic.
In terms of issues and debates, what kind of approach does the use of qualitative data take?
IDIOGRAPHIC.
(A01) What is meant by primary data?
How can primary data be obtained?
Primary data is collected FIRST-HAND by the RESEARCHER directly from a group of participants FOR A SPECIFIC RESEARCH PURPOSE.
The researcher might collect information via observation, psychometric test, interview etc and this data may be qualitative or quantitative.
(A01) What is meant by secondary data?
How does a researcher use secondary data?
When a researcher uses secondary data, this means that SOMEONE ELSE HAS ALREADY COLLECTED THE INFORMATION FOR A DIFFERENT PURPOSE and the INFORMATION HAS BEEN STORED ON RECORD FOR USE BY OTHER RESEARCHERS.
The researcher will re-analyse this second hand data for a new purpose.
Give some examples of types of secondary data?
- Statistical data from the government.
- Medical records of patients with Schizophrenia from a psychiatric hospital that somebody else has gathered.
(A03) What is a limitation of primary data? (think practical issues - is it time consuming?).
P: Primary data can be TIME CONSUMING and EXPENSIVE to conduct.
E: The researcher has to recruit participants to their sample and set up their experiment in a research environment (e.g. a lab). They must also consider ethical guidelines (i.e. the right to withdraw and informed consent). Although research is permitted to be carried out without consent in a public domain, issues arise concerning right to withdraw.
L: This is a limitation of primary data as it is therefore more costly and demanding than accessing pre-existing data from primary sources.
How do practical issues of primary data compare to those of secondary data?
P: Secondary data is a lot MORE TIME-EFFICIENT AND COST-EFFICIENT than primary data.
E: The researcher simply has to consult pre-existing data from previous studies, rather than conducting their own.
E: This is a strength as they do not need to gather research materials or recruit participants/ gather consent to make their research ethically valid as the data already exists in the public domain to be used in further studies.
(A03) Criticism of Secondary Data (think validity compared to primary data).
ii) How does this also link to temporal validity?
P: Secondary data is LOWER IN VALIDITY than primary data.
E: This is because SECONDARY DATA has been collected for a different purpose.
E: This means that FINDINGS MIGHT NOT BE ENTIRELY RELEVANT TO THE RESEARCH QUESTION.
L: Therefore data MAY NOT NEATLY FIT THE NEEDS OF THE INVESTIGATION.
This also links to temporal validity –> LOW temporal validity of secondary data –> Primary data (collected for a different purpose) may have been compiled numerous years ago and may not be generalisable to the target population at the time the secondary research is being conducted. Therefore, means data is not representative of the population today.
Meanwhile, primary data is directly collected by the researcher and has high temporal validity as participants take part in the study DURING THE TIME PERIOD that the data is looking into.
(A03) Why is primary data likely to be reliable? (think in terms of replication - is it easily replicated?)
P: Primary data is likely to be reliable
E: This is because data is collected first-hand. As such, a researcher can carefully plan their research and operationalise variables to measure behaviour.
E: These procedures can be well-documented and carried out in a controlled manner (e.g. following a standardised procedure).
L: This makes replication possible to check for the consistency of research findings.
(A03) Why might replication not be possible for secondary data?
-When reviewing primary data and using it as secondary data, we might not know the procedure the researcher used to conduct their study.
-Without this knowledge of the study’s procedure and methodology, other researchers are not able to replicate the study and check for the consistency of the research findings.
Ethical Considerations - Primary Data - why is it much more difficult compared to secondary data?
P: A limitation of using primary data is that there are many ethical considerations that need to be considered.
E: Primary data involves gathering participants and setting up a research study directly. If the research objective studies potentially sensitive issues such as abnormality and mental illness, it would involve the use of vulnerable participants.
E: Therefore, researchers will need to take care when designing their research not to cause psychological harm to participants and ensure fully informed consent is gained prior to research. Ethical approval for research would also need to be gained from the BPS.
Are there ethical considerations that need to be made when using secondary data?
ii) What do these issues centre around?
iii) In what instance is consent implied?
iiii) When may ethical approval need to be sought?
There are LESS but still ethical issues pertaining to the use of secondary data.
They largely centre around the issue of confidentiality, consent, and safe storage.
If data is in the public domain (e.g. internet source) then consent is implied.
Ethical approval need only be sought if personal information can be used to identify participants or where access to data is restricted.
(A01) What is meta-analysis?
-A QUANTITATIVE research technique that aims to find the results by POOLING data from multiple studies to arrive at ONE COMBINED ANSWER.
-Data from these studies are reviewed together and the combined data can be tested by statistical techniques to assess effect size.
What does meta-analysis integrate results from?
What is the sample size in a meta-analysis?
How does this overcome the issue of small sample sizes?
Integrates results from ALL published studies on the same topic. The sample size is the number of studies regarding the same research question.
This in turn helps overcome the issue of small sample sizes because it combines data/conclusion from lots of smaller studies into one large study, it allows the identification of trends and relationships that wouldn’t be possible from individual smaller studies.
Give an example of when a meta-analysis may be used? What does it generally focus on?
In clinical psychology, examples include the effectiveness of treatments across different patient groups. It generally focuses on effect sizes.
For example, a meta-analysis on the effectiveness of CBT for depression will focus on the size of the effect of CBT found by all of the research gathered.
(A03) Strength of meta-analysis (are they robust and generalisable?)
i) What limitation does meta-analysis overcome (think sample size and does it settle controversies concerning conflicting results)?
ii) How do meta-analysis increase the generalisability of the results from individual studies?
P: Meta-analysis overcomes the limitation of small sample sizes of individual studies, and reduces the risk of false-negative results. Moreover, meta-analysis can settle controversies resulting from studies with conflicting results.
In addition, combining studies with varying sample sizes and patient populations can increase the GENERALISABILITY of the results of individual studies; this allows the results of the meta-analysis to be more representative of a wider population.
(A03) Limitation of meta-analysis (what kind of studies are likely to be published?)
Meta analyses may be impacted by a publication bias.
The problem here is that “positive” studies (where statistical significance has been found) are more likely to be published and subsequently, more likely to be accessed and used within a meta-analysis.
This may lead to inaccurate conclusions being drawn.
What is meant by the ‘file drawer problem’ and how is this a problem with meta-analyses?
” The File Drawer Effect,” coined in 1979 by Robert Rosenthal, refers to the fact that in science many results remain unpublished, especially negative ones.
The exclusion of studies with negative findings may lead to SKEWED REPRESENTATION OF THE AVAILABLE EVIDENCE.
How can this issue with publication bias and the ‘file drawer effect’ in meta-analysis be dealt with?
What may happen if there are too many studies to include all?
Psychologists should locate as many studies as possible that might be suitable for inclusion using as many techniques as possible (e.g., computer and Internet searches, e-mails to active researchers, consulting reference lists, manual searching of related journals).
If there are too many studies to include all, the analyst might randomly sample from the studies or, more commonly, narrow the focus to a meaningful subliterature.
How might researcher bias impact upon the studies selected for a meta-analysis?
How can this be dealt with and improved by meta-analysts?
Researchers often have a tendency to favour studies that align with their preconceived notions or desired outcomes.
This bias may lead to the exclusion of studies with contradictory findings (the ‘file drawer effect’) resulting in a skewed representation of the available evidence.
To improve this, it is important for meta-analysts to employ rigorous and transparent selection criteria and consider a wide-range of studies to minimise the impact of researcher bias.
(A03) ETHICS - Strength of meta-analysis?
However, what is a potential issue with this - is consent always guaranteed?
Meta-analysis is SECONDARY data.
Secondary data already exists in the public domain so researchers make the assumption that informed consent has already been gathered.
However, this is not always guaranteed like it is when primary research is conducted.
(A03) Are meta-analyses scientific in their approach?
How kind of testing can be carried out from data from meta-analysis? Is this an objective way to analyse research findings?
Meta analyses adopt a scientific approach.
They largely gather QUANTITATIVE data to calculate the size of an effect.
Scientific testing can be carried out on the data to look for significance of findings which is an OBJECTIVE way to analyse research findings.
How might reliability be established within a meta-analysis? What should researchers look for when pooling the data from multiple studies?
Researchers should assess the reliability of the data from multiple studies BY REVIEWING THE PROCEDURE and METHODOLOGY of the individual studies.
They should see if a standardised procedure has been followed –> replicability high to check for the consistency of findings.
What do measures of central tendency analyse?
Measures of central tendency analyses how close scores are to the average participant response.
-They are used to summarise sets of numbers, giving a score which is representative of the set.
