The methods you use to analyse your data will depend on whether you have chosen to conduct qualitative or quantitative research, and this choice will be influenced by personal and methodological preference and educational background. It could be influenced also by the methodological standpoint of the person who teaches on your research methods course.

DECIDING WHICH APPROACH TO USE

For quantitative data analysis, issues of validity and reliability are important. Quantitative researchers endeavour to show that their chosen methods succeed in measuring what they purport to measure. They want to make sure that their measurements are stable and consistent and that there are no errors or bias present, either from the respondents or from the researcher.

Qualitative researchers, on the other hand, might acknowledge that participants are influenced by taking part in the research process. They might also acknowledge that researchers bring their own preferences and experience to the project. Qualitative data analysis is a very personal process. Ask two researchers to analyse a transcript and they will probably come up with very different results. This may be because they have studied different subjects, or because they come from different political or methodological standpoints. It is for this reason that some researchers criticise qualitative methods as ‘unscientific’ or ‘unreliable’. This is often because people who come from quantitative backgrounds try to ascribe their methods and processes to qualitative research. This is a fruitless exercise. The two approaches are very different and should be treated as such.

When to analyse data

Quantitative and qualitative data are analysed in different ways. For qualitative data, the researcher might analyse as the research progresses, continually refining and reorganising in light of the emerging results.

For quantitative data, the analysis can be left until the end of the data collection process, and if it is a large survey, statistical software is the easiest and most efficient method to use. For this type of analysis time has to be put aside for the data input process which can be long and laborious. However, once this has been done the analysis is quick and efficient, with most software packages producing well presented graphs, pie charts and tables which can be used for the final report.

ANALYSING QUALITATIVE DATA

To help you with the analysis of qualitative data, it is useful to produce an interview summary form or a focus group summary form which you complete as soon as possible after each interview or focus group has taken place. This includes practical details about the time and place, the participants, the duration of the interview or focus group, and details about the content and emerging themes (see Figures 2 and 3). It is useful to complete these forms as soon as possible after the interview and attach them to your transcripts. The forms help to remind you about the contact and are useful when you come to analyse the data.

There are many different types of qualitative data analysis. The method you use will depend on your research topic, your personal preferences and the time, equipment and finances available to you. Also, qualitative data analysis is a very personal process, with few rigid rules and procedures. It is for this reason that each type of analysis is best illustrated through examples (see Examples 8–11 below).

Formats for analysis

To be able to analyse your data you must first of all produce it in a format that can be easily analysed. This might be a transcript from an interview or focus group, a series of written answers on an open-ended questionnaire, or field notes or memos written by the researcher. It is useful to write memos and notes as soon as you begin to collect data as these help to focus your mind and alert you to significant points which may be coming from the data. These memos and notes can be analysed along with your transcripts or questionnaires.

The qualitative continuum

It is useful to think of the different types of qualitative data analysis as positioned on a continuum (see Fig. 4). At the one end are the highly qualitative, reflective types of analysis, whereas on the other end are those which treat the qualitative data in a quantitative way, by counting and coding data.

For those at the highly qualitative end of the continuum, data analysis tends to be an on-going process, taking place throughout the data collection process. The researcher thinks about and reflects upon the emerging themes, adapting and changing the methods if required. For example, a researcher might conduct three interviews using an interview schedule she has developed beforehand. However, during the three interviews she finds that the participants are raising issues that she has not thought about previously. So she refines her interview schedule to include these issues for the next few interviews. This is data analysis. She has thought about what has been said, analysed the words and refined her schedule accordingly.

Thematic analysis

When data is analysed by theme, it is called thematic analysis. This type of analysis is highly inductive, that is, the themes emerge from the data and are not imposed upon it by the researcher. In this type of analysis, the data collection and analysis take place simultaneously. Even background reading can form part of the analysis process, especially if it can help to explain an emerging theme. This process is illustrated in Example 8.

Comparative analysis

Closely connected to thematic analysis is comparative analysis. Using this method, data from different people is compared and contrasted and the process continues until the researcher is satisfied that no new issues are arising. Comparative and thematic analyses are often used in the same project, with the researcher moving backwards and forwards between transcripts, memos, notes and the research literature. This process is illustrated in Example 9.

Content analysis

For those types of analyses at the other end of the qualitative data continuum, the process is much more mechanical with the analysis being left until the data has been collected.

Perhaps the most common method of doing this is to code by content. This is called content analysis. Using this method the researcher systematically works through each transcript assigning codes, which may be numbers or words, to specific characteristics within the text. The researcher may already have a list of categories or she may read through each transcript and let the categories emerge from the data. Some researchers may adopt both approaches, as Example 10 illustrates.

This type of analysis can be used for open-ended questions which have been added to questionnaires in large quantitative surveys, thus enabling the researcher to quantify the answers.

Discourse analysis

Falling in the middle of the qualitative analysis continuum is discourse analysis, which some researchers have named conversational analysis, although others would argue that the two are quite different.

These methods look at patterns of speech, such as how people talk about a particular subject, what metaphors they use, how they take turns in conversation, and so on. These analysts see speech as a performance; it performs an action rather than describes a specific state of affairs or specific state of mind.

Much of this analysis is intuitive and reflective, but it may also involve some form of counting, such as counting instances of turn-taking and their influence on the conversation and the way in which people speak to others.

The processes of qualitative data analysis

These examples show that there are different processes involved in qualitative data analysis.

• You need to think about the data from the moment you start to collect the information.
• You need to judge the value of your data, especially that which may come from dubious sources.
• As your research progresses you need to interpret the data so that you, and others, can gain an understanding of what is going on.
• Finally, you need to undertake the mechanical process of analysing the data.

It is possible to undertake the mechanical process using computing software which can save you a lot of time, although it may stop you becoming really familiar with the data.

There are many dedicated qualitative analysis programs of various kinds available to social researchers that can be used for a variety of different tasks. For example, software could locate particular words or phrases; make lists of words and put them into alphabetical order; insert key words or comments; count occurrences of words or phrases or attach numeric codes. Some software will retrieve text, some will analyse text and some will help to build theory. Although a computer can undertake these mechanical processes, it cannot think about, judge or interpret qualitative data (see Table 10).

TABLE 10: USING COMPUTERS FOR QUALITATIVE DATA
ANALYSIS: ADVANTAGES AND DISADVANTAGES

ANALYSING QUANTITATIVE DATA

If you have decided that a large survey is the most appropriate method to use for your research, by now you should have thought about how you’re going to analyse your data. You will have checked that your questionnaire is properly constructed and worded, you will have made sure that there are no variations in the way the forms are administered and you will have checked over and over again that there is no missing or ambiguous information. If you have a well-designed and well-executed survey, you will minimise problems during the analysis.

Computing software

If you have computing software available for you to use you should find this the easiest and quickest way to analyse your data.

However, data input can be a long and laborious process, especially for those who are slow on the keyboard, and, if any data is entered incorrectly, it will influence your results. Large scale surveys conducted by research companies tend to use questionnaires which can be scanned, saving much time and money, and you should check whether this option is available.

If you are a student, spend some time getting to know what equipment is available for your use as you could save yourself a lot of time and energy by adopting this approach. Also, many software packages produce professional graphs, tables and pie charts which can be used in your final report, again saving a lot of time and effort.

Most colleges and universities run statistics courses and data analysis courses. Or the computing department will provide information leaflets and training sessions on data analysis software. If you have chosen this route, try to get onto one of these courses, especially those which have a ‘hands-on’ approach as you might be able to analyse your data as part of your course work. This will enable you to acquire new skills and complete your research at the same time.

Statistical techniques

For those who do not have access to data analysis software, a basic knowledge of statistical techniques is needed to analyse your data. If your goal is to describe what you have found, all you need to do is count your responses and reproduce them. This is called a frequency count or univariate analysis. Table 11 shows a frequency count of age.

TABLE 11: AGE OF RESPONDENTS

From this table you would be able to see clearly that the 20–29 age group was most highly represented in your survey. This type of frequency count is usually the first step in any analysis of a large scale survey, and forms the base for many other statistical techniques that you might decide to conduct on your data (see Example 12).

However, there is a problem with missing answers in this type of count. For example, someone might be unwilling to let a researcher know their age, or someone else could have accidentally missed out a question. If there are any missing answers, a separate ‘no answer’ category needs to be included in any frequency count table. In the final report, some researchers overcome this problem by converting frequency counts to percentages which are calculated after excluding missing data. However, percentages can be misleading if the total number of respondents is fewer than 40.

Finding a connection

Although frequency counts are a useful starting point in quantitative data analysis, you may find that you need to do more than merely describe your findings. Often you will need to find out if there is a connection between one variable and a number of other variables. For example, a researcher might want to find out whether there is a connection between watching violent films and aggressive behaviour. This is called bivariate analysis.

In multivariate analysis the researcher is interested in exploring the connections among more than two variables. For example, a researcher might be interested in finding out whether women aged 40–50, in professional occupations, are more likely to try complementary therapies than younger, non-professional women and men from all categories.

MEASURING DATA

Nominal scales

To move beyond frequency counts, it is important to understand how data is measured. In nominal scales the respondent answers a question in one particular way, choosing from a number of mutually exclusive answers. Answers to questions about marital status, religious affiliation and gender are examples of nominal scales of measurement. The categories include everyone in the sample, no one should fit into more than one category and the implication is that no one category is better than another.

Ordinals cales

Some questions offer a choice but from the categories given it is obvious that the answers form a scale. They can be placed on a continuum, with the implication being that some categories are better than others. These are called ordinal scales. The occupationally based social scale which runs from ‘professional’ to ‘unskilled manual’ is a good example of this type of scale. In this type of scale it is not possible to measure the difference between the specific categories.

Interval scales

Interval scales, on the other hand, come in the form of numbers with precisely defined intervals. Examples included in this type of scale are the answers from questions about age, number of children and household income. Precise comparisons can be made between these scales.

Arithmetic mean

In mathematics, if you want to find a simple average of the data, you would add up the values and divide by the number of items. This is called an arithmetic mean. This is a straightforward calculation used with interval scales where specific figures can be added together and then divided.

However, it is possible to mislead with averages, especially when the range of the values may be great. Researchers, therefore, also describe the mode which is the most frequently occurring value, and the median which is the middle value of the range. The mode is used when dealing with nominal scales, for example it can show that most respondents in your survey are Catholics. The median is used when dealing with both ordinal and interval scales.

Quantitative data analysis can involve many complex statistical techniques which cannot be covered in this book. If you wish to follow this route you should read some of the data analysis books recommended below.

SUMMARY

• The methods you use to analyse your data will depend upon whether you have chosen to conduct qualitative or quantitative research.
• For quantitative data analysis, issues of validity and reliability are important.
• Qualitative data analysis is a very personal process. Ask two researchers to analyse a transcript and they will probably come up with very different results.

• After having conducted an interview or a focus group, it is useful to complete a summary form which contains details about the interview. This can be attached to the transcript and can be used to help the analysis.
• Qualitative data analysis methods can be viewed as forming a continuum from highly qualitative methods to almost quantitative methods, which involve an element of counting.
• Examples of qualitative data analysis include thematic analysis, comparative analysis, discourse analysis and content analysis.
• The analysis of large-scale surveys is best done with the use of statistical software, although simple frequency counts can be undertaken manually.
• Data can be measured using nominal scales, ordinal scales or interval scales.
• A simple average is called an arithmetic mean; the middle value of a range is called the median; the most frequently occurring value is called the mode.

FURTHER READING

Qualitative analysis

Dey, I. (1993) Qualitative Data Analysis: A User Friendly Guide for Social Scientists, London: Routledge.

Quantitative analysis