An area of research which has really gained momentum in recent years, qualitative research is often regarded in some sense as competing with the more established (at least in medical circles) quantitative research.
This is unfortunate, since the two approaches should be seen as complementary, providing different perspectives and answering different specific questions within any one broad area.
Quantitative research is concerned with counting and measuring things, producing in particular estimates of averages and differences between groups (e.g., blood pressure of patients treated with two different drugs).
Qualitative research has its roots in social science and is more concerned with understanding why people behave as they do: their knowledge, attitudes, beliefs, fears, etc. (e.g., why do patients prefer to be involved in decision-making about their treatment?)
Qualitative research allows the subjects being studied to give much ‘richer’ answers to questions put to them by the researcher, and may give valuable insights which might have been missed by any other method.
Not only does it provide valuable information to certain research questions in its own right but there is a strong case for using it to complement quantitative research methods.
For example if the area of interest has not been previously investigated then qualitative research may be a vital forerunner to conducting any quantitative research; for example, it’s impossible to carry out a meaningful structured questionnaire survey on patient satisfaction with a service, if the important issues to the patients surrounding the provision of that service are not known.
At the other extreme qualitative research may also help you to understand the findings of quantitative research; for example, it is very easy to discover that some patients fail to keep appointments at outpatients clinics, but uncovering the reasons for this can be more difficult and conventional surveys may miss some of the important factors.
There are three main methods for collecting data in qualitative research. The resulting data is usually transcribed then analysed using one of a variety of techniques for analysis (development [and interpretation] on key themes for example). The three main methods of data collection are:
For this method the researcher brings together a small number of subjects to discuss the topic of interest. The group size is kept deliberately small, so that its members do not feel intimidated but can express opinions freely. A topic guide to aid discussion is usually prepared beforehand and the researcher usually ‘chairs’ the group, to ensure that a range of aspects of the topic are explored. The discussion is frequently tape-recorded, then transcribed and analysed.
Example: Rutman (1996) explored the policy and practice implications of caregivers’ experiences of powerfulness and powerlessness. She used group workshops to generate data. Brainstorming techniques were used to explore the ‘ideal’ caregiving situation.
Data can be collected by an external observer, referred to as a non-participant observer. Or the data can be collected by a participant observer, who can be a member of staff undertaking usual duties while observing the processes of care. In this type of study the researcher aims to become immersed in or become part of the population being studied, so that they can develop a detailed understanding of the values and beliefs held by members of the population.
Sometimes a list of observations the researcher is specifically looking for is prepared before-hand, other times the observer makes notes about anything they observe for analysis later.
Example: Johnson and Webb (1995) used observation to gather evidence about how value judgements made by staff and patients can impact on decision making. In this study, the researcher acted as a participant observer, working as a nurse on the ward while observing situations where nurses were faced with difficult moral choices. Observations were recorded as field notes and analysed for content.
Interviews use the same principle as a focus group, but subjects are interviewed individually, ideally in the patient’s own home. Interviews in qualitative research are usually wide ranging, probing issues in detail. They seldom involve asking a set of predetermined questions, as would be the case in quantitative surveys. Instead they encourage subjects to express their views at length. One particularly useful technique is the critical incident study, in which subjects are asked to comment on real events rather than giving generalisations. This can reveal more about beliefs and attitudes and behaviour. The researcher may be able to obtain more detailed information for each subject, but loses the richness that can arise in a group in which people debate issues and exchange views.
Example: Frederikson, et al (1996) used unstructured interviewing to explore family functioning and interpersonal relationships through the perceptions of women of Vietnam partners in New Zealand. The reasons they give for choosing this method include lack of adequate theory and definitions in the field to produce valid instruments for large-scale survey techniques and the complexity of the social interactions involved in the impact of post-traumatic stress disorder on families.
Further methods used in qualitative research studies
Diary methods – The researcher or subject keeps a personal account of daily events, feelings, discussions, interactions etc.
Role-play and simulation – Participants may be asked to play a role, or may be asked to observe role-play, after which they are asked to rate behaviour, report feelings, and predict further events.
Case-study – This is an in-depth study of just one person, group or event. This technique is simply a description of individuals.
The downside of qualitative research is that, invariably, only small numbers of subjects can be studied because data collection methods are so labour intensive. It is also often criticised for: being subject to researcher bias; the difficulties in analysing qualitative data rigorously; the lack of reproducibility and generalisability of the findings (i.e. findings may not be applicable to other subjects or settings). Proponents of qualitative research would however argue that there are strategies available to the qualitative researcher to protect against these potential biases and to enhance the rigour of the findings. Nicholas and Pope (1995) wrote an article in the BMJ specifically addressing techniques for improving the rigour of qualitative research findings.
The methodological checklist below was developed by the same authors to help readers of qualitative projects assess the quality of published research but they also provide a useful checklist for researchers to consider when designing their own qualitative research.
Check list for the appraisal of qualitative research
– Was the research question clearly identified?
– Was the setting in which the research took place clearly described?
– If sampling was undertaken, were the sampling methods described?
– Did the research worker address the issues of subjectivity and data collection?
– Were methods to test the validity of the results of the research used?
– Were any steps taken to increase the reliability of the information collected, for example, by repeating the information collection with another research worker?
– Were the results of the research kept separate from the conclusions drawn by the research workers?
– If quantitative methods were appropriate as a supplement to the qualitative methods, were they used?
This helpsheet is intended only to provide an introduction to some of the key issues and techniques involved in qualitative research. As mentioned above there are several design issues to consider when conducting this type of research and we would advise new users of this research to seek further information before embarking on your own work. Please contact the R&D Support Unit on 01823 342799.
Johnson, M and Webb, C. (1995) Rediscovering unpopular patients: the concept of social judgement. Journal of Advanced Nursing 21 (3): 466-475
Mays, N and Pope, C. (1995) Rigour and qualitative research. British Medical Journal 311: 109-12
Rutman, D. (1996) Caregiving as women’s work: women’s experiences of powerfulness and powerlessness as caregivers. Qualitative Health Research 6 (1): 90-111
Miles MB and Huberman AM. An Expanded Sourcebook Qualitative Data Analysis – 2nd Edition. 1994
NHS Management Executive. College of Health: Consumer Audit Guidelines. 1994
Qualitative research methods were developed in the social sciences to enable researchers to study social and cultural phenomenon. Examples of qualitative methods include:
action research aims to contribute both to the practical concerns of people in an immediate problematic situation and to the goals of social science by joint collaboration within a mutually acceptable ethical framework;
case study research – a case study is an empirical enquiry that investigates a contemporary phenomenon within its real-life context;
ethnography- the ethnographer immerses her/himself in the life of people s/he studies and seeks to place the phenomena studied in its social and cultural context.
For the research work, the basic RESEARCH QUESTION is factors effecting production planning and control with reference to performance measurement
To address all such questions the methodology used is Qualitative and I make use of Quantitative methods to analyse the statistical data, which to be collected during research work.
QUANTITATIVE AND QUALITATIVE RESEARCH
I am conducting two approaches to investigations my research work i-e. Qualitative & Quantitative In the former, we use words to describe the outcomes and in the latter, we use numbers.
The main methods employed in qualitative research are
Category and concept formation
The generation of theory
In seeking to explore the natural scene, the qualitative researcher aims to be as unobtrusive as possible, so that neither research presence nor methods disturb the situation. This is why participant observation is one of the favoured approaches.
Blends in with natural activity,
Access to the same places, people and events as the subjects,
Documents relevant to the role, including confidential reports and records,
Use of mechanical aids, such as tape recorders and cameras,
First-hand experience of the role and thus heightens understanding of it,
Worthwhile contribution to the life of the institution
The strengths of systematic observation are
It is relatively free of observer bias.
It can establish frequencies, and is strong on objective measures
Reliability can be strong.
Generalise-ability, Once I have devised my instrument, large samples can be covered.
It is precise, There is no ‘hanging around’ or ‘muddling through’.
It provides a structure for the research topic
A great deal of qualitative material comes from talking with people whether it be through formal Interviews or Casual conversations.
It is essential for the researcher:
To develop empathy with interviewees and win their confidence;
To be unobtrusive, in order not to impose one’s own influence on the interviewee.
The best technique for this is the unstructured interview.
There are a number of techniques researchers use in the natural course of the conversation to aid clarity, depth and validity. Here are some:
Check on apparent contradictions
Search for opinions
Ask for clarification
Ask for explanations, pose alternatives
Pursue the logic of an argument
Ask for further information
Aim for comprehensiveness
Put things in a different way
Express incredulity or astonishment
Summarise occasionally and ask for corroboration
Ask hypothetical questions
Play devil’s advocate
The researcher engages in ‘active’ listening, which shows the interviewee that close attention is being paid to what they say; and also tries to keep the interviewee focused on the subject, as unobtrusively as possible. Both kinds of interview might be used in the same research.
Where qualitative research is seeking to generalise about general issues, representative or ‘naturalistic’ sampling is desirable. This covers places, times and persons.
Representative sampling cannot always be achieved in qualitative research because of
a) The initially largely exploratory nature of the research
b) Problems of negotiating access
c) The sheer weight of work and problems of gathering and processing data using only one set of eyes and ears
Documents are a useful source of data in qualitative research, but they have to be treated with care. The most widely used are official documents, personal documents, and questionnaires.
Official documents include registers, timetables, minutes of meetings, planning papers, lesson plans and notes, confidential documents on pupils, school handbooks, newspapers and journals, files and statistics, notice boards, exhibitions, official letters, textbooks, work cards, photographs.
Personal documents are diaries, creative writing exercises, pupils’ ‘rough’ books, graffiti, personal letters and notes.
If these have already been created, they are part of the ‘natural’ situation, and can tell the researcher a great deal about pupil and teacher behaviour, culture and perspectives.
In studies that I have been connected with I have found out a great deal from these kind of documents.
Diaries frequently used in qualitative research. Their very nature speaks to the features outlined in the first section above. They are ‘natural’, they contain personal meanings and understandings, and they are processual.
The researcher needs to know the basis and motivation on which they were compiled.
They are particularly strong, therefore, where used in conjunction with other methods.
Questionnaires are not among the most prominent methods in qualitative research, because they commonly require subjects to respond to a stimulus, and thus they are not acting naturally.
Interaction among techniques in this way is typical of qualitative research.
In order to accord with the features of qualitative research outlined above, one would need to take into account the questions of:-
Questionnaires in qualitative research often contain a mixture of the two.
The need to identify the context in which replies are being given
The need for checks, balances, extensions and modificationsN
Some qualitative researchers are not concerned about validity as it is commonly understood, preferring to aim for ‘understanding’, which might be achieved.Whichever approach one adopts, however, validity or rigour in qualitative research commonly depends on:
The less the researcher disturbs the scene, the longer spent in it, and the deeper the penetration of the research, the more the representation of it might claimed to be authentic.
If we are aiming to understand the meanings and perspectives of those being studied, how better to judge if our understandings are accurate and full than by giving our accounts back to those involved and asking them to judge?
Respondent validation may not always be appropriate or desirable.
The main ethical debates in qualitative research revolve around the tensions between secret and open research, and between the public’s right to know and the subject’s right to privacy.
Participant observation has, on occasions, been likened to ‘spying’ or ‘voyeurism’.
There is a temptation, too, for some researchers to negotiate access into an institution, carry out observations that he or she requires, persuade subjects to ‘spill the beans’, and then ‘cut and run’.
In practical terms, this means, for example, not harming the institution or the persons one is researching, if possible leaving them in a better rather than a worse condition, protecting their identities in disseminating the research.
Respondent validation can be seen to have an ethical dimension.
In qualitative research, analysis frequently takes place at the same time as data collection.
In order to make sense of the data, much may have to be jettisoned – which means a lot of time and work might have been wasted, as well as a lower quality product.
Analysis, therefore, begins almost immediately, with ‘primary analysis’.
Later on, after more data collection in interaction with primary analysis, a second stage occurs with ‘category and concept formation’.
The research might stop at this point, depending on the aims, or it might proceed to a third stage, the ‘generation of theory’.
I shall consider each of these.
As interview transcripts are made, or fieldnotes of observation compiled, or documents assembled,
the researcher continuously examines the data, perhaps highlighting certain points in the text or writing comments in the margins.
These might identify what seem to be important points, and note contradictions and inconsistencies, any common themes that seem to be emerging, references to related literature, comparisons and contrasts with other data and so on..
How did I come across the idea?
As I’m building up numbers of interviews, that is I interview the same person lots of times,
I’ve noticed that they repeat their account of certain incidents, usually fairly important ones in their lives.
The other salient factor is that the account is given in the same words each time, with remarkably little variation.
In addition, this kind of repeating of tales is elicited most often when there has been a gap in my interviewing of a few weeks, so the narrative has gone cold.
They cannot immediately recall exactly what they told me before.
Then I got the repetition of incidents, and the repetition of phrases
Explanations and ideas
It might simply be that the repetition of incidents is due to lapses in memory, especially as people are getting older, that would not be surprising. But there is a problem there, because it fails to explain
Why these incidents should be repeated in exactly the same phraseology?
Why doesn’t the lapse of memory extend to that too?
Why is it that it is only certain things, certain incidents that get repeated?
Category and concept formation
Most qualitative researchers arrive at a point where their data has to be organised in some kind of systematic way, if only for analytic purposes.
It may be helpful to summarise data in some way, tabulate them on a chart, or construct figures, or sketch diagrams. Such distillation helps one to encapsulate more of the material in a glance.
The generation of theory
Many qualitative studies do not go beyond the construction of models and typologies.
This ordered, descriptive detail is a perfectly legitimate pursuit.
As we have seen, it takes considerable work, skill and insight to achieve this level of description, and the results are valuable.
But we might want to go on from asking ‘what’ and ‘how’ questions to ‘why’ questions.
What we saw in the second stage of analysis above was ‘how’ but we would like to know ‘why.’
Types of theory
It is useful to see theories on two dimensions. The first is Glaser and Strauss’s (1967) distinction between substantive and formal theory.
The former is theory that applies to a particular case; formal theory is at a higher level of abstraction and applies to a generality of cases.
The second dimension is that of micro-macro. Qualitative research lends itself more readily to micro research, which is concerned with activity within classrooms and schools, interaction between people, local situations, case studies.
The development of theory proceeds typically through comparative analysis.
As we saw earlier, instances are compared across a range of situations, over a period of time, among a number of people and through a variety of methods.
Comparisons are being made all the time – in checking data, testing an idea, bringing out the distinctive elements of a category, establishing generalities within a group.
Any of these could spark off ideas about ‘why’, which would bring more comparisons to test and refine that idea.
As soon as one begins to identify significant events or words, and goes on to develop categories and concepts, one is building up essential components of theory.
Consulting the literature is an integral part of theory development, and the main way of making comparisons outside the study.
Another important factor is time. The deeper the involvement longer the association, the wider the field of contacts and knowledge
As part of my research, I am looking at certain characteristics (variables) and endeavouring to show something interesting about how they distributed within Production Planning and Control.
A variable needs measured for the purpose of quantitative analysis. Using the data that I have collected then I can make use of
Descriptive statistics including
Variance and standard deviations,
Associations and correlations
Variables can be displayed graphically by tables, bar or pie charts for instance.
This may be all the statistics I need and I can make deductions from my descriptions. In fact, univariate (one variable) analysis can only be descriptive.
However, descriptive statistics used to describe a significant relationship between two variables (bivariate data) or more variables (multivariate).
Infer significant generalise able relationships between variables. The tests employed designed to find out whether or not my data is due to chance or because something interesting is going on.
Mean: is a measure of the central location or average of a set of numbers,
Standard deviation: is the square root of the variance
Variance: is a measure of dispersion (or spread) of a set of data calculated in the following way:
s = âˆ‘ (x – mean)
Median: is the centre or middle number of a data set
Quartiles: divide a distribution of values into four equal parts. The three corresponding values of the variable are denoted by q1, q2 (equal to the median) and q3
Range: is a measure of dispersion equal to the difference between the largest and smallest value.
Measures of Location and Dispersion
A distribution is symmetrical if the difference between the mean and the median is zero.
An appropriate pictorial representation of the data, (histogram, stem and leaf diagram etc.) would produce a mirror image about the centre:
A distribution is positively skewed (or skewed to the right) if the mean – median is greater than zero. Such data when represented by a histogram would have a right tail that is longer than the left tail
A distribution is negatively skewed (or skewed to the left) if the mean – median is less than zero. Such data when represented by a histogram would have a left tail that is longer than the right tail
If data skewed then the best measure of location is the median and the best measure of dispersion is the inter-quartile range. If data are symmetrical then the best measure of location is the mean and the best measure of dispersion is the standard deviation or variance.
This is an important concept in statistics and is an important part of our story.
It is defined in the following way: if an experiment has n equally likely outcomes and q of them are the event E, then the probability of the event E, P(E), occurring is
Testing an hypothesis
There are two basic concepts to grasp before starting out on testing an hypothesis.
Firstly, the tests are designed to disprove hypotheses. We never set out to prove anything; our aim is to show that an idea is untenable as it leads to an unsatisfactorily small probability.
Secondly, the hypothesis that we are trying to disprove is always chosen to be the one in which there is no change. For example there is no difference between the two population means.
This is referred to as the null hypothesis and is labelled H0. The conclusions of a hypothesis test lead either to acceptance of the null hypothesis or its rejection in favour of the alternative hypothesis H1.
Hypothesis testing: a hypothesis test or significance test is a rule that decides on the acceptance or rejection of the null hypothesis based on the results of a random sample of the population under consideration.
Step 1: Formulate the practical problem in terms of hypotheses.
Step 2: Calculate a statistic that is a function purely of the data.
Step 3: Choose a critical region.
Step 4: Decide the size of the critical region.
In hypothesis testing, the t test is used to test for differences between means when small samples are involved. For larger samples use the z test. The t test can test
If a sample has been drawn from a Normal population with known mean and variance.
If two paired random samples come from the same Normal population.
Any hypothesis test can be one tailed or two tailed depending on the alternative hypothesis, H1.
Consider the null hypothesis, H0: m =3
A one tailed test is one where H1 would be of the form m > 3.
A two-tailed test is one where H1 would be of the form m ¹ 3.
Single sample test
Let X1, X2, ¼ , Xn be a random sample with mean and variance s2. To test if this sample comes from a Normal population with known mean m and unknown variance s2,
T = X – µ
S âˆ•âˆšn -1
The test statistic used to test the null hypothesis H0: the population mean equals m.
If the test statistic lies in the critical region whose critical values are found from the distribution of Tn, a, H0 is rejected in favour of the alternative hypothesis H1. n are the degrees of freedom and for a single sample test n = n-1, and a is the significance level of the test.