WIXLIB

MATRIX QUERIES AND MATRIX DATAREPRESENTATIONS IN NVIVO 12 PLUSNVivo Advanced(Updated)

OVERVIEW1. Matrices and their basic structures2. Types of elements (variables) for matrix comparisons3. Setting up matrix queries in NVivo 12 Plus4. Specific matrix “use cases” in qualitative, quantitative, and mixed methodsresearch5. Wrap-up2

IF MULTILINGUAL SELECTION OF A BASELANGUAGE (“TEXT CONTENT LANGUAGE”)Go with default (determinedby QSR International basedon purchaser location in theworld)ORNVivo ribbon - File tab - Info - Project Properties - General (sub) tab - Textcontent language (dropdownmenu)3

1. MATRICES AND THEIRBASIC STRUCTURES4

DEFINITION(S) AND PURPOSESGeneral PurposesStructureMatrices are used to explore and showdata relationships and patterns“a rectangular array of quantities orexpressions in rows and columns that istreated as a single entity and manipulatedaccording to particular rules” (in amathematics sense; in a computer science /information technology sense) (per Googlesearch engine) In NVivo, any defined element may be used topopulate the respective matrices NVivo uses (intensity) matrices to report out onsentiment analysis autocoding findings NVivo 12 Plus uses (intensity) matrices to reportout on theme and subtheme autocoding findingsa complex of lines intersecting at rightangles (via rows and columns)5

MATRICES ARE ABOUT RELATIONSHIPSTypes of Common Relationships found in Matrices: hierarchical such as general to specific, topic to subtopic,within the set of, part of a type, contained within, found withinincluded with, co-occurring with,associated with (some relationship but not fully defined)appearing in close proximity with,similar to, related by likenessdifferent from (orthogonally unrelated)6

MATRICES ARE ABOUT RELATIONSHIPS(CONT.)For example, some relationships described by matrices include: Which source documents (articles) are related to particular themes? Which themes and subthemes are related to which documents? Which terms from text sets are related to the four sentiment categories (very negative, moderatelynegative, moderately positive, and very positive)? What are similarities and dissimilarities between the coding of two coders (in a dyadic comparison in acoding comparison)? What are macro themes in a coding structure or codebook (through a matrix coding query)?7

TYPES OF MATRICESMatrices are referred to by their application (matrix type), which includes the typesof data and the analytical uses of that data Confusion matrix / contingency table or cross-tabulation analysis / error matrix (predicted values vs.actual empirical values) These are sometimes used to highlight the differences between Type I and Type II errors in basic signals detection Effects matrix Sentiment analysis intensity matrix (a temperature matrix, to a color-saturation heatmap but inmatrix format or with intensity indicated by number) Relational or network matrices (for relationships), and othersMatrices are generically referred to sometimes by the numbers of their elements intheir rows and columns (as in a 5 x 7 matrix, or a 2 x 2 matrix)8

TYPES OF MATRICES(CONT.)Specific topical matrices are referred to by their main contents (one content typealong its column headers and the other along its rows) For example: type-by-document matrices, document-by-theme matrices, and others9

GENERAL FEATURESMatrices do not have to be symmetrical in terms of the labels on Column A1 and Row1A (in the next diagram)The numbers of entities (column and row headers) in the rows and columns do nothave to match; they do not have to pair either (but may depending on the type ofmatrix) Data may be incomplete, and matrices still have informational value (they are robust even in thecondition of missing data)Tables tend to be more structured, with unique records on each row (runninghorizontally) and variables at the top of each column (running vertically)Matrices may be automatically extracted (by computer); they may be manuallycreated10

BASIC STRUCTUREMatrix Variables (Columns)A1Matrix Variables (Rows)Data Cells11

MATRIX INDICATORS(W/ DEFINED ROW AND COLUMN LABELS/HEADERS)Variables may be Of a similar kind or type (in the matrix): all nodes / case nodes (individuals to individuals, ororganizations to organizations) / codes; all interviews (in various groupings); all (cross-referenced)responses to questions, etc. Of mixed kinds or types (across rows and columns, not within the cells): themes and researchdocuments; physical locations and interview subjects; themes (terms) and interviews; themes (concepts)and categorical outcomes; individuals and organizations; time periods and themes; sources in NVivo(sources, memos, codes, transcripts, interviews, and others)In the data cells Presence or non-presence of a relationship (1 or 0; a binary finding) Frequency of occurrence of relationship; “strength” of relationship (may be turned into a networkgraph); intensity Contents of matrix variable overlap in text (content) format; coded text12

MATRIX CELLSCells come at the intersection of respective column headers and rows (individualrecords)Matrices may be labeled “sparse” if there are more cells with 0s than with nonzeroes (whether in binary matrices or strength-of-relationship matrices)13

IN QUALITATIVE AND MIXED METHODSRESEARCH, MATRICES Are word- or text-based; may include a quantitative aspect (usually frequency countsas an indicator of relationship strength) Text may be “raw” (primary source data; transcripts) or highly processed (edited research articles)Are based on variables (nodes, themes, interviewee / survey taker / focus groupparticipant demographics and “characteristics” for grouping, and others)May be used at various scales: the micro- (cell-level), meso- (relational, dyadic,triadic / motif ), and macro- (matrix-scale pattern)May be designed (1) based on a targeted question, (2) based on the need to surfaceleads for further exploration (such as a “text summarization” application), (3) basedon pure exploratory discoveryMay contain single or multiple queries14

2. TYPES OF ELEMENTS (VARIABLES)FOR MATRIX COMPARISONS15

VARIABLES FOR MATRIX COMPARISONSIN NVIVO 12 PLUSAny text (at the most atomistic level)Any attributes or variablesAny groups of text or multimediarepresented by text descriptors (foldersof contents)Any categorical variablesAny codes (nodes, case nodes)Any groups of nodesAny “coded by” set (of codes / nodes) Such as indicated by classification sheetsAny relationshipsAny modelsAny model items Coder-based comparisons16

THE RESEARCH PROCESSGeneral OverviewPossible Matrix ApplicationsLiterature reviewText summarization (themes and documents)Research designRelevant document identification (for close reading)in a literature reviewResearch instrument prototyping and design (oracquisition) and pilot testingSamplingPattern identification in interviews, surveys, notes,and codesResearchData visualization as relational network graphs (foranalysis, for presentation)Data collection and othersData cleaningData analysisWrite-up and presentation17

SOME TYPES OF “ASKABLE” QUESTIONSWITH MATRICESAre there instances of particular text in particular “searchable” documents (.PDF, .txt,.rtf, .doc, .docx, etc.)?Are there locational or spatial patterns in (textual) data?Are there temporal patterns in (textual) data?Are there topical or theme patterns in (textual) data?Are there similarities / differences between responses of individuals from differentdemographic or categorical or spatial or other groupings? (from interview, survey,focus group, or other similar types of data)Are there relationships between concepts? Individual entities? Group entities?18

3. SETTING UP MATRIX QUERIESIN NVIVO19

DATA INGESTION IN PROPER FORMSSizes of ContentsAll text ingested atomistically (the lowest indivisible “unit of analysis” or “record” ormulti-media-based “object” such as an article or a memo) If a series of articles are all created as one text set, generally, the text set will be queried as onedocument (instead of a series of articles in a text corpus)Accessible Machine-Readable TextAll scanned text as “searchable” or optical character recognition (OCR) textAll video transcribed into machine-readable textAll audio transcribed into machine-readable textAll imagery alt-texted into machine-readable text20

DATA PREPARATIONGeneral in NVivoSpecific in NVivoAll relevant research materials included andcoded; clear data labeling, consistentnaming protocolsCombined master file of group codedprojects, with multiple user codedcontents (for runs of interrater reliability)Various types of groupings (by folder, bynode, by nickname, by classification variable,and others), without creating dataredundancy (which skews text queries andtext frequency counts and other types ofanalyses); may delete redundant text fordata queries or create new (sub) NVivoprojects with select data for particular dataqueriesRelationships defined and linkedCase node source classifications appliedModels createdGoal: all data fully exploited in clear ways21

PATHS USED TO CONDUCT DATA QUERIESRESULTING IN DATA MATRICESNVivo ribbon - Query tab - Matrix CodingNVivo ribbon - Query tab - Coding ComparisonNVivo ribbon - Query tab - Group Query (models, relationships, attributes, codingat)( to live and interactive demos)22

QUERIES FOLDER STORAGEMatrix queries are storedin the Queries folder (unless saved elsewhereby the researcher)23

RELATED DATA VISUALIZATION: MATRIX “CHART”24

A RELATIONAL MATRIX TO A NETWORK GRAPH(READ ACROSS)AABCDEFGHIBCDEFGHI----------25

A RELATIONAL MATRIX TO A NETWORK GRAPH(CONT.)Node-link diagram (vertex-relationship diagram)26

PATHS USED TO CONDUCT AUTO CODING RESULTING ININTENSITY DATA MATRICES (IN NVIVO 12 PLUS)Theme and subtheme extraction / topic modeling Highlight source. (You can use CTRL A to select all in a folder.)In ribbon, select Analyze tab. Click Auto Code Button.Select “Identify themes ”Proceed with the Auto Code Wizard Sentiment extraction Highlight source. (You can use CTRL A to select all in a folder.)In ribbon, select Analyze tab. Click Auto Code Button.Select “Identify sentiment ”Proceed with the Auto Code Wizard 27

28

29

30

31

4. SPECIFIC MATRIX “USE CASES” In Qualitative and MixedMethods Research32

MACHINE-READING RESEARCH ARTICLES (OR OTHERTEXTS) FOR THEMES AND SUB-THEMES(TO SAVE ON HUMAN “CLOSE READING,” ESP. OF RELATIVELY “BIG DATA” CORPUSES USING AUTOCODING)Theme or Conceptor Phenomena orIndividual(keyword orphrase)Theme or Conceptor Phenomena orIndividualTheme or Conceptor Phenomena orIndividualTheme or Conceptor Phenomena orIndividualResearch article(or source) #1“““““33

COMPARING AND CONTRASTING RESEARCH SUBJECTRESPONSES BY CATEGORICAL GROUPINGSSexAge riable34

COMPARING AND CONTRASTING RESEARCH SUBJECTRESPONSES BY (CATEGORICAL) OUTCOMESOn-time GraduationLate le35

EXPLORING POTENTIAL LOCATIONAL OR SPATIALPATTERNSLocation #1Location #2Location #3Location #4Interview Subject#1 / or Nodes / or etc.#2#3 36

EXPLORING POTENTIAL TIME PATTERNS (LIKE CHANGESOVER TIME, LIKE PRE-POST EVENT TIME CHANGES)Time Period 1Time Period 2Time Period 3Time Period 4VariableVariableVariable 37

IDENTIFICATION OF OVERLAPS IN CODIFIEDTHEMES (MATRIX CODING QUERIES)NodeNodeNode NodeNodeNode 38

OUTLIER CASE COMPARISONSOutlier Case AOutlier Case Z39

COMPARING SENTIMENTS, EMOTIONS, ATTITUDES,AND BELIEFSSentimentsEmotionsAttitudesBeliefs Interviewee #1 40

CROSS-QUERY RESPONSE ANALYSIS(COMPARISONS AND CONTRASTS)Question 1ResponsesQuestion 2Responses41

COMPARING CHANGES ACROSS TIME PERIODS(TYPES OF TIME: DISCRETE, PERIODIC, CONTINUOUS; SHORT-TERM VS. LONGITUDINAL)Time Period #1Time Period #2 Variables of aType42

INTER-RATER RELIABILITY(SIMILARITY/DIFFERENCE ANALYSIS)Coder A CodingCoder B Coding43

AUTOMATED SENTIMENT ANALYSISPositiveVery veVery NegativeTweetstreamsInterviewsFacebook PostingsSurvey Responses Organizedby TopicNewspaper Articles on aSpecific TopicAd Hoc #Hashtag DiscussionsOthers 44