A researcher was interested in learning about the reasons that people chose to engage in high-risk recreational activities. The sparse literature that exist

74

MODULE 6: WHAT THIS INFORMATION TELLS YOU:

DEPICTING and ANALYZING the DATA

LESSON 1: ANALYZING QUALITATIVE DATA

Qualitative data is data recorded as language rather than numbers. Qualitative data are typically

collected over time in the form of systematic field notes, interview notes and transcripts, researcher

journals and reflections, and archival records. These data can then be conceptually organized,

beginning with the identification of preliminary patterns while the researcher is still in the field.

Qualitative data require careful analyses that employ much the same logic of finding and matching

patterns that dominates quantitative methods in order to reveal patterns and principles that distinguish

the phenomenon under study. Qualitative methods require rigorous inquiry on which consumers of the

research can rely when drawing conclusions. The design and data collection principles provide the

foundation for the inquiry, but, like the raw numbers recorded in a computer data file, they have no

inherent knowledge value until they have been carefully analyzed. Many beginning researchers using

qualitative methods underestimate the discipline and hard work necessary to turn interesting data into

qualitative research through analysis. Collecting data in the field is often fascinating, even fun;

analyzing the data systematically once the researcher has left the field is time-consuming and

sometimes tedious, particularly if done manually. The rule of thumb that qualitative researchers often

use to plan the time they need to complete a qualitative research analysis is: “Time in the field equals

time in analysis.”

The steps qualitative researchers use to analyze their data involve three basic steps: a) preparing

and organizing the data for analysis, 2) exploring the data, and 3) describing and developing themes

from the data. As a whole, the data analysis process is primarily inductive – going from the particular

(the detailed data consisting of transcriptions or typed notes from interviews, and the like) to the

general – codes and themes. The process includes both simultaneous and iterative (cycling back and

forth) phases. In qualitative research, the data collection and analysis are simultaneous activities. This

is one way in which the analytic process in qualitative research analysis differs from that in

quantitative research. In the latter, the process of data collection is separate from, and necessarily

precedes, the data analysis. In qualitative research inquiries, you need to collect and analyze data to

determine what questions to ask in an emerging design. The phases are also iterative, where the

researcher cycles back and forth between data collection and analysis. The process of data analysis in

qualitative research can be visually depicted to illustrate these steps and characteristics, as follows:

75

The researcher collects data

(i.e., a text file, such as field notes,

transcriptions, or scanned materials)

iterative

The researcher prepares the data

for analysis (i.e., transcribes field notes)

The researcher reads through the data

(i.e., obtains a general sense of the material)

The researcher codes the data

(i.e., locates text segments and

assigns a code label to them)

simultaneous

Codes the text for description to Codes the text for themes to

be used in the research report be used in the research report

Preparing and Organizing the Data for Analysis

Initial preparation of the data for analysis requires organizing the vast amount of information,

transferring it from spoken or written words to text, and making decisions about how to analyze the

data. At an early stage in qualitative analysis, researchers organize their data into computer files, file

folders, or index cards or notes. Organization of the data is critical in qualitative research because of

the large amount of information gathered during a study. With a sizable amount of data, transcribing

and organizing information requires a system of organization. This organization could take several

forms, such as: developing a matrix or a table of sources that can be used to help organize the

material; organizing the material by type, such as all interviews, all observations, all documents, all

photographs or other visual material, or by participant, site, location, or some combination of any of

these; keeping duplicate copies of all forms of data.

Qualitative data often consists of words to be analyzed by interviewing participants or recording

field notes or obtaining other materials (letters, diaries, records, etc.). These words need to be

converted to a form for data analysis. Transcription is the process of converting recordings or field

notes into text data. A transcriptionist may be employed for this purpose, and there are special

76

machines that can help create a transcript. Some helpful suggestions have been offered for

transcription that will facilitate the analysis:

a) create 2-inch margins on each side of the text document so that you can jot down notes in the

margins during data analysis

b) leave extra space on the page between the interviewer’s comments and the interviewee’s

comments. This enables the reader to clearly distinguish between speakers during data

analysis

c) highlight or mark in some way the questions asked by the interviewer. These questions will

not be analyzed, and identifying them clearly can allow the analyst to skip over them

d) use complete, detailed headers that contain information about the interview or observational

session. The headers in the interview and observational records show the type of content to be

included in the transcription

e) transcribe all words, and type the word [pause] to indicate when individuals being interviewed

take a lengthy break in their comments. These pauses may provide useful information about

times when interviewees cannot or will not respond to a question. Other actions occurring

during an interview can also be noted. For example, type [laughter] when the interviewee

laughs; [telephone rings] to indicate a phone call that interrupts the interview; or [inaudible] to

mark when it cannot be determined what was being said. As a general approach, transcribing

all words will provide data that captures the details of an interview.

Exploring the Data

After the data have been organized and transcribed, data analysis begins. The first step in data

analysis is to explore the data by reading through all of the information to obtain a general sense of it.

A preliminary exploratory analysis in qualitative research consists of obtaining a general sense of the

data, memoing ideas, thinking about the organization of the data, and considering whether more data

is needed. Writing memos in the margins of field notes, transcripts, or under photos helps in this

initial process of exploring the data. These memos are short phrases, ideas, concepts, or hunches that

occur to you, and they help form an initial analysis of the data.

Describing and Developing Themes from the Data

Beyond having a general understanding of the data, it is critical to address the research question.

This involves a process of examining the data in detail to describe what you learned and developing

themes or broad categories of ideas from the data. Describing and developing themes from the data

consists of answering the research questions and forming an in-depth understanding of the

phenomenon under study. Description is a central feature of reporting what you learn about a cultural

group and the setting in which it lives and interacts – the focus of ethnographic research. On the other

hand, thematic development plays a central role in grounded theory, where you generate themes or

categories of information from the data. A narrative study typically includes both – a description of

the events in an individual’s story and the themes that emerge from that story. What is most important

at this time is how you proceed from a database to generating a description, a theme, or both for a

qualitative study. Several procedures are used in this process:

• coding text (or image) data

• developing a description from the data

77

• defining themes from the data

• connecting and interrelating themes

Coding Data

The process of analyzing text or images in qualitative research begins when the researcher codes

the data. Coding is the process of segmenting and labeling text to form descriptions and broad themes

in the data. Although there are no set guidelines for coding data, some general procedures do exist. A

diagram like the one below will help you to visualize and understand this process.

Initially read Divide the text Label the segments Reduce overlap Collapse codes

through the into segments of information and redundancy into themes text data of information with codes of codes

Many pages Many segments 30-40 Codes reduced Codes reduced of text of text codes to 20 to 5-7 themes

The object of the coding process is to make sense out of text data, divide it into text or image

segments, label the segments, examine codes for overlap and redundancy, and collapse these codes

into themes. Thus, this is an inductive process of narrowing data into a few themes. Several steps are

involved in coding data. Although there is no definitive procedure, the following steps are

recommended:

1. Get a sense of the whole. Read through all of the transcriptions carefully. Jot down in the

margins some ideas as they come to mind.

2. Select one document (e.g., one interview, one field note). Choose the most interesting, the

shortest, or the one on the top of the pile. Go through it, asking the question, “What is this

about?” Consider the underlying meaning and write it down in the margin in two or three

words, drawing a box around it.

3. Begin the process of coding the document. This involves identifying text segments, placing a

bracket around them, and assigning a code word or phrase that accurately describes the

meaning of a text segment. Sentences or paragraphs that all relate to a single code are called a

text segment. Codes are labels used to describe a segment of text or an image. Codes can

address many different topics, such as the following:

78

• setting and context

• perspectives held by participants

• participants’ way of thinking about people and objects

• processes

• activities

• strategies

• relationships and social structure

The codes can be stated in the participants’ actual words which are called in vivo codes. They

can also be phrased in other terms more conversant to the researcher, or expressed in the

language of the observer (e.g., a researcher’s statement about “individuals talking to each

other”).

4. After coding an entire text, make a list of all code words. Cluster together similar codes and

look for redundant codes. The objective is to reduce a long list of codes to a smaller, more

manageable number, such as 25 to 30. It is best not to over-code the data because, in the end,

you will need to reduce the codes to a smaller number of themes.

5. Take this list and go back to the data. Try out this preliminary organizing scheme to see

whether new codes emerge. Circle specific quotes from participants that support the codes.

6. Reduce the list of codes to get 5 to 7 themes or descriptions of the setting or participants.

Themes are similar codes aggregated together to form a major idea in the database. In the

literature you might find that themes are also called categories (e.g., in grounded theory

research), dimensions, issues, or perspectives. Regardless of the name, identify the 5-7 themes

by examining codes that are most frequently discussed by participants, are unique or

surprising, have the most evidence to support them, or those that might be expected from

studying the phenomenon. The reason for the small number of themes is that it is best to write

a qualitative report providing detailed information about a few themes rather than general

information about many themes. A description is a detailed rendering of people, places, or

events in a setting in qualitative research.

7. From the coding and the themes, construct a narrative description and possibly a visual display

of the findings for your research report.

Below is a sample transcript from an interview and the codes, themes, and ideas that the researcher

recorded in the margins. The transcript resulted from an interview from a project exploring residents’

perceptions of some changes they were told would likely occur in the recreational facilities that are

open to them during weekend hours.

Notice that the margins on the left and right are extra wide so that the researcher could jot down ideas

there.

On the left side codes are inserted; on the right side, ideas and emerging themes. These features could

have been reversed (e.g., codes on the right) depending on the preference of the researcher.

For codes, two or three words were used as labels. These were the actual words used by the

participant, “CL.” Sometimes codes are shown as possible alternatives.

79

schools

gangs

crime and trouble

pranks

Boxes are placed around key words that the researcher wanted to use as codes or possibly themes.

By bracketing (drawing a bracket around sentences in the left margin), the researcher identified

sentences that seemed to “fit together” to describe one idea, the text segments. Notice that not all

sentences are bracketed; the goal was to reduce the text to 5 to 7 themes for the final research report.

The researcher asked herself, “What are they talking about here?”, drew a bracket around key

sentences that related to a code, and assigned a code label.

On the right side, potential themes in the transcript are noted. Also on the right side are the early

jottings of the researcher, and a good quote that they might want to use in the final report.

Codes Here Themes (and

other Ideas) here

LB: Are you aware of the recent proposal to drastically

reduce the number of hours that the rec centers will be

open, Mrs. Connell? What do you think about that?

Will that likely have an impact on you or your family?

What about the community?

CL: Well, I had heard about it – saw the article in the

paper and heard it on the 10:00 news. I have to tell you,

emotion: mad it made me pretty mad, but I guess I see the reasons why. Potential theme:

school probs There’s so much written about the problem in the rec activities

gangs, fights with and fighting and all – and I think if kids deter negative

participated more in sports and activities after school behaviors

better social together, they would learn to get along and there wouldn’t

relations be so many problems. If the rec centers will be closed,

fewer, probs where will the kids go? Will they be hanging out on the Potential theme:

streets more? And that surely means trouble – you know, rec activities

hanging out you get a group of kids together and they do a lot of develop

more crime and want to impress each other, and outdo each other. So I positive behaviors

think closing the rec centers much earlier will results in more

for our kids. And what about the kids that

lose income are employed part-time there? They’ll lose income that might Potential theme:

prob for family be important for them and their families. You know, we hear impact on family

about how kids are getting fatter and lazier these days – all the

junk food and video games and TV – so how does closing the

rec center help? I think it’s gonna tell kids to stay home and

poor health watch even more TV and play more computer games and get

80

helpful

sedentary less fresh air and more passive at home. And that means it’s Idea: impact is

obesity gonna have a big effect on their health – more reason to be in in home, school,

hyperactivity bad health. I have a son who’s kind of hyperactive at school – streets, community

gets in trouble a lot for getting out of his seat and walking

rec activities help around the room and fidgeting. I think it’s been for him Theme: Impact

sleeping to go to the gyms after dinner and an hour. It helps him sleep on child w/probs

concentration and it helps him concentrate so he can do his homework. And

I even think he’s becoming a better basketball player so maybe

h.s. athletics when he gets to high school he can get on the team. And then Idea: Could help

less drinking he’ll hang around with better kids who

less drugs so much. So I’m not happy at all with what they’re planning on

more not less! doing – I want the rec centers to be open

in future

Good quote!

The process now, beyond this single transcript section, will be to continue the coding and theme

development so that the researcher can construct a description of the community and write about

themes relating to this change in the availability of recreational facilities.

Using Codes to Build Description

Now that you know the general process and have looked at an actual example of coding a

transcript, we go on to explore how to use codes to build description and themes in qualitative

research. Since description is a detailed rendering of people, places, or events in a setting in

qualitative research, it is easiest to start the analysis after the initial reading and coding of the data. In

some forms of qualitative research design, such as in ethnography or in case studies, the researcher

provides considerable description of the setting Developing detail is important, and the researcher

analyzes data form all sources (e.g., interviews, observations, documents) to build a portrait of

individuals or events. To describe an event, the analyst might ask, What occurred in this setting?”

Alternatively, the researcher might describe an individual by asking, “What is this person like?” For

describing a place, the question might be, “What is this place like?” The description needs to be so

detailed that the reader is transported to a research site or can visualize a person.

Using Codes to Develop Themes

In addition to description, the use of themes is another way to analyze qualitative data. Because

themes are similar codes aggregated together to form a major idea in the database, they form a core

element in qualitative data analysis. Like codes, themes have labels that typically consist of no more

than two to four words. Through initial data analyses, it is likely that 30-50 codes can be identified. In

subsequent analyses, these codes are reduced to 5-7 major themes through the process of eliminating

don’t drink and do drugs

more hours, not less!

81

redundancies and codes that cannot be conveniently categorized. There are several types of themes,

and these are typically identified as:

• Ordinary themes: themes that a researcher might expect to find

• Unexpected themes: themes that are surprises and not expected to surface during a study

• Hard-to-classify themes: themes that contain ideas that do not easily fit into one theme

or that overlap with several themes

• Major and minor themes: Themes that represent major ideas, or minor, secondary ideas,

in a database

In the preceding example of coding a transcript, a major theme might be: benefits of recreation. Minor

themes might be: delinquency, crime, social relationships, illicit substances, special needs.

One more aspect of theme development is important to note. A realistic presentation of

information does not present only one side or the other. In an attempt to capture the complexity of

situations, qualitative researchers actively analyze data for contrary evidence. Contrary evidence is

information that does not support or confirm the themes and provides contradictory information about

a theme.

One final important point about developing themes is that you will reach a point where a theme is

fully developed and new evidence will not provide additional insight. Saturation is the point where a

theme is developed and detailed and no new information can add to its specification. Unquestionably,

the point at which you achieve saturation is a judgment call, but most qualitative researchers are able

to realize when this point occurs. In the development of the theme of “benefits of recreational

participation,” the researchers drew on their extensive interviews with community members. They

found evidence through specific examples and quotes to illustrate the theme, and when they returned

to the interview transcripts, no new information surfaced during their reading and re-reading of the

transcripts. They concluded that they had reached saturation on this theme. A check with participants

(recall “member checking” as a form of validating qualitative methods and findings) confirmed that

they had adequately specified this theme.

Connecting and Interrelating Themes

Many qualitative studies include only description and themes for analysis. However, by adding

the layering of themes or interconnecting them, you can build sophistication and complexity into the

research. Thus, qualitative data analysts represent this complexity by layering and interconnecting

themes.

Layering themes builds on the idea of major and minor themes, but organizes the themes into

layers from basic elementary themes to more sophisticated ones. Layering the analysis (also called

first and second order abstractions) means representing the data using interconnected levels of themes.

Minor themes are subsumed within major themes and major themes lead to broader themes. The

entire analysis becomes more complex as the researcher works upward toward broader and broader

levels of abstraction. The number of layers may vary from two to four or five, and recognizing these

layers will help you read more complex qualitative studies. For example, consider the figure below

about the study of reactions to the proposed change in recreational facilities. This figure shows that

the authors used four layers, including the database as the first layer. The authors collected data from

several sources (layer 1) and analyzed it to develop a description of the reactions (layer 2). From this

description they then formed four themes (layer 3) and combined these themes into two broad

82

Social Psychological

perspectives (layer 4). Knowing how layering works helps you see how the authors began with the

details and worked their way up to the more general themes in their analysis.

Two broad perspectives

Four themes identified from the data

Impact on Impact on Recreation

Emotion Self Family Benefits

Descriptive analysis of the Description of Events

chronology of events

following the announcement

Database: Interview transcriptions, Data

newspaper reports, videotapes from

news segments, field notes from

observations in recreation centers

A second thematic analysis approach interconnects the themes. Interconnecting themes means that

the researcher connects the themes to display a chronology or sequence of events, such as when

qualitative researchers generate a theoretical and conceptual model as in grounded theory research, or

report stories of individuals, such as in narratives. Often qualitative researchers draw a visua