Qualitative vs. Quantitative research methods PART

Qualitative research strategies: Observations
Interviews
Cases studies These often involve face-to-face interactions beteen researcher and participant The researchers need to be flexible and sensitive to the needs of the social context within which the data is obtained. Step 1 Step 2 The data is then analysed and interpreted. To look for themes is more common than trying to confirm a hypothesis.<br>

Qualitative – words and anlysis
Qualitative – numbers- easy to summarize and use in statistics. Meant for generalization beyond the sample from which the data is drawn. Gathered through direct interaction with participants. Open-ended and flexible ”rich data”<br>

When dealing with qualitative research – it is imperative to be able to tolerate a degree of uncertainty.<br>

According to qualitative research: ”Researchers can only come to understand the social world through participants’ interpretations – interpretative approach.
Reality is diverse and multifaceted. The goal is to get a picture of this reality. To measure means to reduce it – and therefore lose meaning.<br>

Historical flashback: Quantitative methods, such as the experiment, have been used partly in order to maintain the appearance of psychology as a scientific discipline with valid knowledge claims.
During the 20th century there was a shift away from seeing quantitative methods as the only valid way of gaining data – but also a realization that both methods are needed.<br>

What decides whether to use qualitative vs. Quantitative data? Purpose of research
Characteristics of participants
Researchers’ beliefs about the nature of knowledge and how it can be aquired (see next slide)<br>

The nature of knowledge – Ritchie and Lewis questions based on the epistenological discussion (how can we know abou tthe world) What is the relationship between the researcher and the researched? Can the researcher be objective. Can the researched ever behave naturally. No? Well, then reflexibility is needed.
What can be held as truth? Accurate measures (natural sciences) or by being supported by something else (social sciences)
How is knowledge gathered? Deductive (cause and effect, generalization and prediction) vs. Inductive (collected evidence used to reach a conclusion – focus: to understand the process).<br>

Rolfe: Means that the distinction between qualitative and quantitave research is a textbook creation and that there is no unified qualitative paradigm.
In fact, he claims, they are not separated.<br>

Quantitative methods assume:

That variables can be identified (and operationalised)
That The relationship between variables can measured by statistics.
Realibility and objectiveness is seen as highly important (and possible) – therefore controlled environments (like labs) are preferred.
Aim: to infer a cause-effect relationship and to be able to generalize from the study.
Eg. experiments and correlational studies. Qualitative methods assume: Exercise 1: try to fill in what qualitative research has in common opposed to quantitative methods based on what we have covered so far:<br>

PART II EVALUATION OF QUALITATIVE RESEARCH METHODS<br>

Evaluation: strengths Provide rich data – that is, in-depth descriptions of individual experiences.
Particularly useful for investigating complex and sensitive issues.
Explain phenomena – that is, go beyond mere observation to understand what lies behind them (eg. why do people become homeless?)
Generate new ideas and theories to explain and overcome problems.
People are studied in their own environment, which increases credibility.<br>

Evaluation: limitations Can be very time-consuming and generate a huge amount of data.
Data analysis can be difficult because of the amount of data and no clear strategy for analysis.
Interpretation of data may be subjective (but reflexivity can help to minimize this)<br>

Is it possible to generalize from qualitative data? This is often the aim of research, but not always so for qualitative research.
Representative generalization – can the findings be applied to populations outside the population of the study? Samples are often small and not selected for being statistically representative so this makes generalization difficult. However, if evidence from other studies confirms the findings (confirmability through eg. triangulation) it is argued that generalization is possible (Hammersley, 1992)
Inferential generalizability – same thing but with the difference that it is the setting of the research that is to be generalized to other settings. Transferability. Depends on the depth of the description of the context – and this may allow for inferences to be made – but needs to be supported or disproved by further evidence (e.g. transferability check through triangulation)
Theoretical generalizability – if the theoretical concepts can be used to open up new fields and develop further theory.<br>

Criteria for judging quantitative and qualitative research Credibility Internal validity
”Trustworthyness” How believable are the research conclusions? Conclusions and interpretations are
Breadth and deapth is gathered. correct as variables are well defined and measures well controlled.
Transferability Generalizability
The context is well described as it is unlikely that The research conclusions can be applied to
it won’t have an impact on the findings. Different samples as the research context is
controlled enough.
Dependability Reliability
Data obtained cannot be expected to be the same Repeated use of the instrument provide stable
Dependability means therefore that the researcher has measurements and researchers using them
Described all factors that might have influenced the data. Find similar results
Confirmability Objectivity
Sujectivity is not only unavoidable; it is valued. Therefore
researchers should give details of procedures and attempt As many sources of bias from opinion are
To find examples that contradict the findings. Eliminated from the research process.<br>

A study is trustworthy if, and only if, the reader of the reseach report judge it to be so” (Rolfe, 2006)<br>

Something on triangulation As a way to increase credibility, but also check transferability, dependability and confirmability triangulation is often used.

Triangulation = a cross-checking of information and conclusions in research, brought about by the use of multiple procedures or sources. If there is agreement between these, there is support of the interpretation of data.

Using triangulation does not mean you get a certain truth, but you get closer to it – reflexivity is still necessary.<br>

Example of triangulation techniques: Method triangulation. Comparing data that come from the use of different methods. These could be both quantitative and qualitative. Eg. first using a questionaire to ask about eating habits in a school, and then conduct focus group interviews afterwards.
Researcher triangulation – involves using different people as researchers. This increased the confirmability and credibility of conclusions. Without this data collection and conclusions might be affected by researcher bias.
Other triangulation thechniques include data triangulation and theory triangulation.<br>

Reflexivity Refers to the researcher’s need to constantly be aware of how and why they are conducting the research, and to recognize at what points their own beliefs and opinions might have influenced data collection or analysis.

To undergo an interview with collegues is a way to expose possible bias. Examiner’s hint: to answer a question about the value of reflexivity in qualitative research, you should make reference to the different opportunities for reflexibility provided by interviews, case studies and observations.<br>

EFFECTS OF PARTICIPANT EXPECTATIONS AND RESEARCHER BIAS Participant expectations – the participants’ ideas of the researcher and the research which can affect the trustworthiness of the data. Pleasing the experiment (or the screw you effect).
Researcher bias – the researcher does not pay enough attention to the participants. This leads to the result that it is the researcher’s own beliefs that determine the research effect.
Can be checked through interviews, credibility checks and reflexivity.<br>

PART III SAMPLING METHODS<br>

Sampling methods The sampling numbers are generally smaller thani n quantitative research.
To get random, representative samples from target population is seldom possible (or the aim) of qualitative research. It is not intended to be statistically representative.
Instead, a sample is chosen because it represents important characteristics of a population – characteristics that are the main concern in evaluation of research. Sampling methods in qualitative research differs from those used in quantitative research.<br>

Examples of sampling in qualitative research Purposive sampling – the participatns are chosen on the basis of particular characteristics that will help the researcher to explore the research topic. Eg. specific experiences, social roles etc. It may be important that there is a diversity – but this is not necessary nor at times possible. Problem – the sample might be biased but this is thought to be lessened if the criterias on which participants were chosen are clearly documented.<br>

Snowball sampling The researcher simply asks participants in the study if they know any other potential participants.
+ time and cost efficient.
+ can be used to get hold hidden populations
- Will most probably lead to biased samples.
- Ethics: confidentiality concerns. Convenience samples can also be used.<br>

Exercise IIYou have been asked to conduct a qualitative research study on football fans’ perception of their favourite team. What sampling method would be appropriate? Why?
Could another sampling method be considered? Why?
What should you consider overall when selecting your sample?<br>

PART IV ETHICS<br>

Ethics In large these are the same as in quantitative research (informed consent, protection from harm, respect for the participants’ integrity and privacy and right to withdraw).
Special here is to be open to problems linked to the private nature often researched, that the researcher might get personally involved and lose objectivity.
Specifically in case studies – anonymity issues (case study with covert observation – no consentform or right to withdraw – problematic).<br>

Learning outcomes – answer these: Distinguish between qualitative and quantitative data.
Explain strenths and limitations of a qualitative approach to research
Explore the extent to which findings can be generalized from qualitative studies.
Explain the importance of credibility in qualitative research.
Explain the effect of triangulation on the credibility/trustworthiness of qualitative research
Explain reflexivity in qualitative research.<br>