Qualitative Data AnalysisSoftware

A workshop for staff & studentsSchool of PsychologyMakerere University

Julius F. Kikooma (PhD) January 27, 2016

Outline for the workshop

2

CAQDAS NVivoOverview

Practice

Julius F. Kikooma

CAQDAS

Before we start…

What is qualitative data?

What are some of the examplesof qualitative data sources?

What is qualitative analysis?

4Julius F. Kikooma

What is qualitative data?

Non-numerical – converse of quantitativedataTypically word based – but may includeimage, video, etc.Can record attitudes, behaviours,experiences, motivations, etc.Descriptive – describing events/opinions etc.Explanatory – explaining events/opinions etc.

5Julius F. Kikooma

Examples of Qualitative Data Sources

6

InterviewsFocus groupsSpeechesQuestionnairesJournals/diaries

DocumentsObservationAudio/visualmaterialsWebsitesSocial media

Julius F. Kikooma

Analysing Qualitative Data

7

Identifysimilarities

Extract themes

Identifyrelationships

Highlightdifferences

Creategeneralisations

Julius F. Kikooma

Qualitative data analysis

The identification, examination and interpretation ofthemes in the data to answer research questions.Miles and Huberman (1994), see qualitative dataanalysis as involving data reduction, data display,and drawing conclusions - a process parallel toquantitative analysis.It is in this context that most CAQDAS hasdeveloped.CAQDAS instead allows the researcher to operateon an entirely new level.

8Julius F. Kikooma

What is CAQDAS?

Computer Aided Qualitative Data AnalysisSoftware (CAQDAS)A database with some powerful qualitativeanalysis toolsCAQDAS searches, organizes, categorizes,and annotates textual and visual data.Programs of this type usually support theory-building through the visualization ofrelationships between data and/or theoreticalconstructs.

9Julius F. Kikooma

CAQDAS tools

Main features to handle the datainclude:

Content searchingLinking toolsCoding toolsQuery toolsWriting and annotation toolsMapping or networking tools

10Julius F. Kikooma

Examples of CAQDAS

QSR NVivo[http://www.qsrinternational.com]ATLAS.ti [http:www.atlasti.com],MAXqda [http:www.maxqda.com]

11Julius F. Kikooma

Organizing Data for analysis

developing yourcodes

coding your data

finding themes,patterns, andrelationships

summarizing yourdata.

12Julius F. Kikooma

Coding is a process for categorizingyour data.Develop a set of codes using bothcodes that you predefine and onesthat emerge from the data.Predefined codes are categories andthemes that you expect to see basedon your prior knowledge.

Developing your codes

13Julius F. Kikooma

Divide data into meaningful unitsUse words/phrases e.g. ‘physical environment’,‘interpersonal relationships’

Codes can be ‘data-driven’ or ‘theory-driven’A priori codes are developed before examining thedataIn vivo codes are derived from the dataCo-occurring codes partially or completely overlapIn NVivo, codes are stored within Nodes

Keep a master list of codes used

Coding Data

14Julius F. Kikooma

• What are theattributes of thesource?

Descriptive

• What are thetopics beingdiscussed?

Thematic• What is going on?• How can this be

interpreted?

Analytic

Types of Code

15Julius F. Kikooma

Types of Code

This took place atHead OfficeThis is aboutdiscriminationagainst womenThis is a reflectionon misogyny in theworkplace

Analytic

Descriptive

Thematic

16Julius F. Kikooma

Step back from the detailed work of codingyour data and look for the themes, patterns,and relationships that are emerging acrossyour data.Look for similarities and differences in

different sets of data and see what differentgroups are saying.

Finding themes, patterns, andrelationships

17Julius F. Kikooma

A theme is generated when similar issues andideas expressed by participants within qualitativedata are brought together by the researcher intoa single category or cluster.This ‘theme’ may be labelled by a word orexpression taken directly from the data or by onecreated by the researcher because it seems tobest characterize the essence of what is beingsaid.

Themes

18Julius F. Kikooma

Write a summary of what you are learning afteryou have coded a set of data, such as transcriptsof interviews or questionnaire responses,.Summarize the key themes that emerge across aset of interview transcripts. Include quotations thatillustrate the themes.Look across the various summaries andsynthesize your findings across multiple datasources.

Summarizing your data

19Julius F. Kikooma

NVivo Overview

Qualitative Analysis Using NVivo

Import Code Query &Visualise

AnnotateSummarise

Julius F. Kikooma21

Importing Sources

Name documents appropriatelybefore importingText-based data can be imported in .doc(x), .rtf, .txtor text-based .pdf format

For Microsoft Word documents, apply consistent headingstyles to use autocoding

Multimedia files can be imported in a variety offormats including: .mp3/4, .wav, .jp(e)g

Edit videos before importing Julius F. Kikooma 22

Other Datasets

Can connect to SurveyMonkey to import surveyresultsImport datasets such as Excel spreadsheets orAccess database tables

Cannot edit datasets after importing – format andstructure datasets before importing

Use NCapture to import social media data such asFacebook, Twitter or LinkedIn feeds

Julius F. Kikooma23

Coding in NVivo

• Descriptive code• Classification/attributeWhat is this?

• Thematic code• Annotation/memoWhy is this

interesting?

• Analytic code• Memo

Why is this relevantto my research

question?

Julius F. Kikooma 24

Creating Nodes

Use a separate node for each elementWho, what, how, whenEach node should encompass one concept only

Text can be coded at multiple nodesMove free nodes into trees where appropriateOrganise trees based on conceptual relationships

Not observed or theoretical associationsE.g. events, strategies, attitudes, beliefs, characteristics

Each concept should appear in only one treeJulius F. Kikooma

25

Tree Structures in NVivo

Julius F. Kikooma 26

Queries

Find and analyse words or phrasesText Search Query – search for a word/phrase

Create a word tree

Word Frequency Query –most frequently occuring words

Create a tag cloudUse memos to record what you learn

Julius F. Kikooma27

Literature Reviews in NVivo

Create a source folder called ‘Literature’Code articles by themes

Create nodes for statistics, quotes, definitions, etc.Annotate content you want to follow-upUse memos to add descriptions or critiquesUse source classifications for date, author, etc.Use queries to find common themes or gaps

Julius F. Kikooma 28

Practice

How confident do you feel…?

Analysing qualitativedataNavigating NVivoCreating a projectAdding data sourcesCreating a node treeCoding deductivelyCoding inductively

Using classificationsUsing setsUsing search foldersCreating chartsCreating tree mapsCreating graphsRunning a text search queryRunning a word frequencyqueryRunning a matrix coding query

30Julius F. Kikooma