USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

USE OF COMPUTER IN DATA

ANALYSIS
Comprehensive Exam Preparation Notes
Topics: Data Visualization • Qualitative Data Analysis Using Computers

SECTION 1: DATA VISUALIZATION

Definition: Data visualization is the graphical representation of information and data using visual
elements such as charts, graphs, maps, and dashboards. Computers have made it possible to
handle large datasets and produce accurate, interactive, and publication-quality visualizations
quickly.

Why Computers are Used for Data Visualization:

• Handle very large datasets that cannot be managed manually
• Produce precise and reproducible outputs
• Allow dynamic updates when data changes
• Enable interactivity (e.g., zooming, filtering, drilling down)
• Facilitate collaboration and sharing via digital formats

1.1 Types of Charts and Graphs

A. Bar Chart (Bar Graph)
Description: A chart that presents categorical data with rectangular bars. The height or length of
each bar is proportional to the value it represents.
Types: Vertical bar chart, Horizontal bar chart, Grouped bar chart, Stacked bar chart.
Best Used For: Comparing discrete categories or groups.

Example: Bar Chart

A researcher studying exam performance across five departments (Sciences, Arts,
Commerce, Law, Engineering) uses a vertical bar chart in Microsoft Excel to compare
mean scores. The x-axis shows departments; the y-axis shows mean score (0–100).
A glance reveals that Sciences scored highest and Law the lowest.

B. Line Graph
Description: Displays information as a series of data points connected by straight line segments.
Ideal for showing trends over time (time-series data).
Best Used For: Showing change over time; tracking trends.

Example: Line Graph

A public health analyst plots monthly malaria cases in a district from January to
December using SPSS graph editor. The rising line from October to December

Page 1 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

clearly signals a seasonal surge aligned with the rainy season.

C. Histogram
Description: A histogram is a graphical display of data using bars of different heights. Unlike a
bar chart, a histogram groups numerical data into ranges (bins/class intervals), and each bar
represents the frequency or relative frequency of values within that range. Bars in histograms
touch each other (no gaps) because the data is continuous.
Feature Bar Chart Histogram
Data Type Categorical (discrete) Numerical (continuous)
Bars Gaps between bars Bars touch (no gaps)
X-Axis Categories/groups Numerical ranges (bins)
Purpose Compare groups Show distribution of data
Example Sales by department Distribution of ages in a sample

Example: Histogram
A lecturer collects the ages of 200 students and uses SPSS to generate a histogram.
Bins are set at: 16–18, 19–21, 22–24, 25–27, 28+.
The tallest bar at 19–21 shows that most students fall in that age range.
The shape is approximately bell-shaped (normal distribution).

Key Histogram Concepts:

• Class interval / Bin width: the range each bar covers (e.g., 5 years)
• Frequency: number of data points falling in each bin
• Relative frequency: proportion of data in each bin (frequency ÷ total)
• Shape of distribution: symmetric, skewed left, skewed right, bimodal

Exam Tip – Histogram Shapes

Normal (bell-shaped): data clusters around the mean. Right-skewed (positive skew): tail
extends to the right; mean > median. Left-skewed (negative skew): tail extends to the left;
mean < median. Bimodal: two peaks, suggests two sub-populations.

D. Pie Chart
Description: A circular chart divided into slices to illustrate numerical proportion. Each slice
represents a category's contribution to the whole (100%).
Best Used For: Showing part-to-whole relationships; proportions.
Limitation: Difficult to interpret when there are many categories; avoid for more than 6 slices.
Example: Pie Chart
A market researcher uses Excel to display the market share of four mobile brands:
Samsung (35%), Tecno (28%), Infinix (20%), Others (17%).
The pie chart instantly shows Samsung dominates the market.

E. Scatter Plot (Scatter Diagram)

Page 2 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Description: A plot that displays values for two variables as points in a two-dimensional space.
Used to observe and show relationships (correlation) between variables.
Best Used For: Exploring relationships, detecting outliers, testing correlation.
Example: Scatter Plot
A nutrition researcher plots daily calorie intake (x-axis) against body weight (y-axis)
for 50 participants using STATA. A positive upward trend suggests a positive correlation.
Outliers (individuals with high intake but low weight) are immediately visible.

1.2 Dashboards
Definition: A dashboard is a visual display of multiple data visualizations (charts, tables, KPIs,
maps) consolidated onto a single screen or page to provide an at-a-glance view of key information
and performance indicators.

Characteristics of a Good Dashboard:

• Concise: only the most relevant metrics
• Interactive: allows filtering, drilling down, and selecting date ranges
• Real-time or near-real-time data updates
• Visually clean: consistent colors, minimal clutter
• Contextual: includes benchmarks, targets, or comparisons

Software Used for Dashboards:

Software Type Typical Use Case
Microsoft Excel Spreadsheet / BI Academic, business, and government reporting
Tableau Dedicated BI tool Advanced interactive dashboards, data
exploration
Power BI (Microsoft) Business Intelligence Corporate reporting, connecting to databases
SPSS Statistical software Academic research and analysis
R (Shiny, ggplot2) Programming language Custom research dashboards
Python (Matplotlib, Programming language Data science and scientific research
Plotly)
Google Data Studio Cloud-based BI Marketing and web analytics
(Looker)

Example: Research Dashboard

A hospital quality manager creates a Power BI dashboard with four panels:
• A line graph: monthly patient admissions over 12 months
• A bar chart: admissions by ward (Pediatric, Maternity, Surgery, Medical)
• A KPI card: current bed occupancy rate (78%)
• A heat map: peak admission hours by day of the week
A single screen reveals operational patterns without reading a lengthy report.

1.3 Interpretation of Outputs

Page 3 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Generating a chart or table is only half the task. A researcher must correctly interpret and
communicate what the visual shows. Below are guidelines for interpreting common outputs:

Interpreting a Bar Chart:

• Identify the highest and lowest bars (range of values)
• Note the scale on the y-axis to avoid misinterpretation
• Compare groups and describe differences in plain language
• Mention any surprising findings or outliers

Interpreting a Histogram:
• Describe the shape (normal, skewed, bimodal, uniform)
• Identify the modal class (bin with the highest frequency)
• Comment on spread (narrow = less variability; wide = more variability)
• Note any gaps or unusual peaks

Interpreting a Line Graph:

• Identify the overall trend (increasing, decreasing, fluctuating, stable)
• Point out the highest and lowest values (peaks and troughs)
• Note the rate of change (steep = rapid change; gradual = slow change)
• Link trends to real-world events where possible

Interpretation Example: Line Graph on Inflation

Figure 3 shows monthly inflation rates from January to December 2023.
The rate rose steadily from 8.2% in January to a peak of 14.7% in July,
before declining to 9.3% by December. The sharp rise between March and July
corresponds with rising fuel prices following supply chain disruptions. The
subsequent decline may reflect tightening monetary policy by the central bank.

1.4 Report Writing with Tables and Figures

In academic and professional reports, tables and figures (charts, graphs, maps, photographs)
must be properly formatted and referenced in text.

Rules for Tables in Reports:

• Every table must have a number and a descriptive title placed ABOVE the table
• Columns and rows must have clear headers with units of measurement
• Source of data must be indicated below the table
• Tables must be referenced in the text before they appear (e.g., “as shown in Table 2”)
• Do not present a table and then ignore it – always interpret key findings

Rules for Figures (Charts/Graphs) in Reports:

• Every figure must have a number and a descriptive caption placed BELOW the figure
• Axes must be labelled with variable names and units
• A legend must be included when multiple categories are plotted
• Figures must be cited in the text before they appear (e.g., “see Figure 1”)
• Avoid unnecessary 3D effects that distort interpretation

Numbering Convention:
Element Caption Position Example Label

Page 4 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Table Above the table Table 1: Mean exam scores by gender

Figure (chart, graph) Below the figure Figure 2: Distribution of ages of
respondents
Histogram Below the figure Figure 3: Histogram of household
income distribution
Photograph Below the figure Figure 4: Field observation site, Abuja,
2023

Sample Table in a Research Report

Table 3: Frequency Distribution of Respondents by Level of Education

Level of Education | Frequency | Percentage (%)

----------------------|-----------|---------------
No formal education | 45 | 15.0
Primary | 60 | 20.0
Secondary | 105 | 35.0
Tertiary | 78 | 26.0
Postgraduate | 12 | 4.0
Total | 300 | 100.0

Source: Field Survey, 2024

Interpretation: Table 3 shows that the majority of respondents (35.0%) had

secondary education, while only 4.0% held postgraduate qualifications.

SECTION 2: QUALITATIVE DATA ANALYSIS USING

COMPUTERS
Definition: Qualitative data analysis (QDA) is the process of examining and interpreting non-
numerical data to understand concepts, opinions, experiences, and social phenomena.
Computers are increasingly used to manage, organize, and analyse qualitative data efficiently.

2.1 Nature of Qualitative Data

Qualitative data is non-numerical and is collected to gain an in-depth understanding of human
experience, behaviour, attitudes, and social contexts. It answers questions of ‘how’, ‘why’, and
‘what does this mean’ rather than ‘how many’.

Key Characteristics of Qualitative Data:

• Rich, detailed, and descriptive in nature
• Context-dependent: meaning depends on the context in which it was collected
• Subjective: reflects the perspective of the participants
• Flexible: can evolve as new themes emerge during analysis
• Non-standardised: no fixed response categories

Page 5 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

2.2 Types / Forms of Qualitative Data

A. Text Data
Text is the most common form of qualitative data and includes written documents, responses, and
records.
Sources:
• Open-ended questionnaire responses
• Written diaries, journals, and field notes
• Government documents, policy papers, newspapers
• Social media posts, emails, letters, and online forums
• Historical records and archival documents

Example: Text Data

A researcher studying youth unemployment collects open-ended survey responses.
One respondent writes: “I have applied to over 30 jobs in the past year. Employers
always say I lack experience, but how do I get experience if no one hires me?”
This text reveals themes of frustration, a catch-22 employment barrier, and
structural challenges in the job market.

B. Transcripts
Definition: A transcript is a written verbatim record of spoken language — a word-for-word
conversion of audio or video recordings into text.
Types:
• Verbatim transcript: includes every word, pause, filler ("um", "uh"), laughter
• Edited transcript: minor editing for readability; removes false starts and repetitions
• Intelligent verbatim: removes fillers and irrelevant content but keeps meaning

Software Used for Transcription:

• [Link] – AI-powered automatic transcription
• Microsoft Word (Dictate feature) – live transcription
• Express Scribe – professional transcription software
• [Link] – human and AI-based transcription services

Example: Interview Transcript (Excerpt)

Researcher: How did the community respond to the new health policy?
Participant (P1): [sighs] Honestly... people were confused. Nobody explained it
properly. Even the health workers at the clinic, they didn’t
seem sure themselves. So people just ignored it.

Analysis Note: Themes emerging – ‘poor communication’, ‘lack of trust in health

workers’, ‘community disengagement’.

C. Interview Data
Definition: Data collected through direct interaction between a researcher and a participant(s).
Interviews may be structured, semi-structured, or unstructured.

Type Description Computer Use

Page 6 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Structured Pre-set questions; fixed order; no Responses entered in

deviation spreadsheets or survey tools
Semi-structured Guide questions but flexible probing Recordings transcribed using
allowed transcription software
Unstructured (in- Open conversation; researcher follows Full recordings transcribed; coded
depth) the participant's lead in NVIVO/[Link]
Focus Group Group discussion; 6–12 participants Audio/video recorded and
transcribed for analysis

Example: Focus Group Interview

A public health study uses focus group discussions (FGDs) with community leaders
to understand barriers to child vaccination. The 90-minute session is recorded on
a digital recorder, transcribed verbatim using [Link], and then imported into
NVivo for thematic coding. Themes identified include: ‘fear of side effects’,
‘religious objections’, and ‘distrust of government programmes’.

D. Observations
Definition: Observation data is collected by the researcher directly watching, experiencing, and
recording behaviour, events, or social phenomena in their natural setting.
Types of Observation:
• Participant observation: researcher immerses in the group being studied
• Non-participant observation: researcher observes without involvement
• Structured observation: uses a pre-designed checklist or schedule
• Unstructured observation: open-ended, narrative field notes

Computer Support for Observation Data:

• Field notes typed directly into tablets or laptops using tools like Evernote or OneNote
• Video observation recorded and later coded using NVIVO or Transana
• Behavioural observation using dedicated software: Observer XT, BORIS
• GPS and mobile apps to log time, location, and activities

Example: Classroom Observation

An education researcher conducts structured observation of teaching methods in
10 secondary school classrooms. A pre-designed observation checklist is entered
into a tablet app. The researcher codes each 5-minute interval: lecture (L),
group work (G), individual work (I), or discussion (D). Data is exported to
Excel and a pie chart is generated showing that 68% of time was spent in
lecture mode across all observed lessons.

E. Multimedia Data
Definition: Qualitative data that exists in non-text formats including audio recordings, video
recordings, photographs, and digital media content.

Multimedia Type Examples Analysis Software

Page 7 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Audio recordings Interviews, oral histories, radio Audacity, NVivo, [Link]

broadcasts
Video recordings FGD sessions, documentary footage, Transana, NVivo, ELAN
classroom observation
Photographs Community mapping, artefacts, NVivo, [Link], MAXQDA
visual elicitation
Social media content Tweets, YouTube videos, Facebook NCapture (for NVivo), Brandwatch
posts
Drawings/artwork Children’s drawings, participatory MAXQDA, manual coding
diagrams

Example: Video Data Analysis

A child development researcher records children playing in a playground.
Videos are imported into NVivo. The researcher codes specific behaviours:
‘cooperative play’, ‘solitary play’, ‘conflict’, ‘sharing’. NVivo allows video
clips to be tagged directly, and a query shows that cooperative play is
significantly more frequent in the 6–8 age group compared to 4–5 year olds.

2.3 Computer-Assisted Qualitative Data Analysis Software (CAQDAS)

CAQDAS refers to software tools designed to help researchers manage, code, and analyse
qualitative data. They do NOT analyse data automatically — the researcher still makes
interpretive decisions. The software simply organises and retrieves data efficiently.

Software Full Name Key Features

NVivo NVivo Qualitative Data Coding, querying, matrices, word frequency,
Analysis social media import, mapping
[Link] Archive for Technology, Network views, code grouping, multimedia, geo-
Lifeworld, and Language mapping
MAXQDA MAX Qualitative Data Analysis Mixed methods, visual tools, focus group
analysis, MAXMaps
Dedoose Dedicated Qualitative Analysis Web-based, team collaboration, mixed methods,
excerpts
QDA Miner Qualitative Data Analysis Text retrieval, coding frequency, integration with
Miner WordStat

2.4 Process of Qualitative Data Analysis Using Computers

1. Data Preparation: Transcribe audio/video; organise documents; import files into CAQDAS
2. Familiarisation: Read all data thoroughly to gain an overview; write analytical memos
3. Coding: Assign codes (labels/tags) to meaningful segments of data
4. Categorisation: Group related codes into broader categories
5. Theme Development: Identify overarching themes from categories
6. Interpretation: Explain what the themes mean in relation to the research question

Page 8 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

7. Report Writing: Present findings with illustrative quotes, tables, and matrices

Types of Coding:
• Open coding (initial coding): first pass; label anything interesting
• Axial coding: identify relationships between codes
• Selective coding: select the core categories central to the research
• In vivo coding: use the participant's own words as the code label

Key Exam Distinction

Quantitative analysis asks: 'How many? How much? Is there a significant difference?'
Qualitative analysis asks: 'What does this mean? Why does this happen? What is the
experience like?' Computers assist both, but the tools and outputs are fundamentally
different.

Example: Thematic Analysis Using NVivo

Research Question: What factors influence adherence to antiretroviral therapy (ART)?
Data: 20 in-depth interviews with HIV-positive patients (transcribed verbatim).
Process in NVivo:
Step 1 – Import 20 transcripts as text files into NVivo.
Step 2 – Read through all transcripts; note initial impressions in memos.
Step 3 – Code using open coding: e.g., ‘side effects’, ‘stigma from family’,
‘cost of transport’, ‘forgetfulness’, ‘support from spouse’.
Step 4 – Group into categories: Barriers to adherence; Enablers of adherence.
Step 5 – Run Word Frequency Query in NVivo – confirms ‘stigma’ appears 87 times.
Step 6 – Final themes: (1) Psychosocial barriers, (2) Economic barriers,
(3) Health system barriers, (4) Social support as enabler.
Reporting: Themes are presented with verbatim quotes and a coding matrix table.

2.5 Reporting Qualitative Findings

Qualitative findings are reported differently from quantitative results. Key conventions include:

• Use verbatim quotes to illustrate themes — quotes give voice to participants

• Identify quotes with participant code (e.g., P3, FGD2-M) to protect anonymity
• Use tables to present coding frameworks, theme summaries, or participant profiles
• Use matrices (cross-tabulation of themes by participant groups) to compare findings
• Provide thick description: explain context, not just content
• Integrate theory to explain why themes emerged

Reporting Element Purpose Example

Verbatim quote Illustrate and validate a theme "They treated us like we were criminals
with the participant’s own words just for being poor." (P7, female, 34)
Coding matrix table Show which codes/themes Table showing themes across 5 FGDs
appeared in which data sources
Frequency table Show how often a code Table 4: Code frequency across 20
appeared across transcripts interviews

Page 9 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Conceptual Illustrate relationships between Figure 5: Conceptual framework of ART

map/diagram themes adherence barriers
Participant profile table Describe the characteristics of Table 1: Profile of interview participants
informants

2.6 Advantages of Using Computers for Qualitative Analysis

• Efficiency: large volumes of data can be managed without physical files or index cards
• Auditability: coding decisions are recorded and can be reviewed (enhances rigour)
• Retrievability: specific codes, phrases, or themes can be retrieved instantly
• Collaboration: multiple researchers can code the same project and compare (inter-rater
reliability)
• Mixed methods: some tools (e.g., MAXQDA, Dedoose) integrate qualitative and
quantitative data
• Visualization: word clouds, concept maps, and matrices are generated automatically

2.7 Limitations of Using Computers for Qualitative Analysis

• No automated meaning-making: computers do not interpret; the researcher must still think
deeply
• Learning curve: CAQDAS tools require time and training to use effectively
• Risk of coding obsession: over-reliance on codes may lead to losing the holistic picture
• Cost: professional CAQDAS licences (e.g., NVivo) can be expensive
• Loss of immersion: over-reliance on software may distance the researcher from the data

SECTION 3: SUMMARY COMPARISON TABLE

Aspect Quantitative Data Visualization Qualitative Data Analysis
(Computer-Assisted)
Data Type Numerical, structured Text, audio, video, images
Primary Tools Excel, SPSS, Tableau, R, Python NVivo, [Link], MAXQDA
Output Charts, graphs, tables, dashboards Codes, themes, matrices, concept
maps
Analysis Process Statistical computation Coding, categorising, interpreting
Key Skill Statistical interpretation Critical, reflective interpretation
Visual Outputs Histograms, bar charts, line graphs Word clouds, concept maps, matrices
Rigour Indicators Validity, reliability, significance Credibility, transferability,
dependability

SECTION 4: EXAM TIPS AND LIKELY QUESTIONS

Page 10 | Research Methods & Data Analysis

USE OF COMPUTER IN DATA ANALYSIS | Exam Preparation Notes

Question Type Likely Question Key Points to Cover

Definition/Explanation What is a histogram? How does it Definition, continuous vs categorical,
differ from a bar chart? no gaps, bins, shape
Application Describe how you would use a Choice of chart (histogram), software
computer to present survey data on (Excel/SPSS), steps, interpretation
income distribution.
Critical Analysis Discuss the advantages and CAQDAS benefits + limitations; role of
limitations of using computers in researcher’s judgement
qualitative data analysis.
Comparison Distinguish between quantitative and Data type, tools, process, outputs,
qualitative approaches to data rigour
analysis.
Report Writing Explain the rules for presenting Numbering, captions, position,
tables and figures in a research referencing in text, sources
report.
Examples Give an example of multimedia Video/audio; CAQDAS tool; coding
qualitative data and explain how it is process
analysed.

Final Revision Checklist

1. Know the difference between a histogram and a bar chart (very common exam question).
2. Be able to name and describe at least 5 types of charts with examples. 3. Know what a
dashboard is and name 3 software tools used to create them. 4. Know the 5 forms of
qualitative data: text, transcripts, interviews, observations, multimedia. 5. Know at least 3
CAQDAS tools and their features. 6. Understand the qualitative analysis process: prepare
→ read → code → categorise → theme → interpret → report. 7. Know the rules for tables
(title above) and figures (caption below) in reports.

End of Exam Preparation Notes

Good luck in your examination

Page 11 | Research Methods & Data Analysis