analysis of quantitative and qualitative data

3.2 Market Research – Data Analysis

1. Why Conduct Market Research?

Market research supplies the evidence a business needs to make informed strategic decisions. It helps to:

• Identify the size and growth rate of a market – informs market‑entry or expansion decisions.
• Understand customer profiles, needs and buying behaviour – guides product development, positioning and pricing.
• Analyse competitors and the competitive environment – shapes competitive strategy and differentiation.
• Test new product concepts, pricing, promotion and distribution ideas – reduces the risk of costly launch failures.
• Reduce risk by basing decisions on evidence rather than intuition – supports investment appraisal and budgeting.

2. Types of Data

• Quantitative data – numerical information that can be measured (e.g., sales figures, percentages, frequencies).
• Qualitative data – non‑numerical information that reveals attitudes, motivations and opinions (e.g., interview transcripts, open‑ended questionnaire responses, observations).

3. Primary vs. Secondary Research

Aspect	Primary Research	Secondary Research
Definition	Data collected first‑hand for the specific research project.	Data already published or collected for another purpose.
Typical Sources	Surveys, interviews, focus groups, observations, experiments.	Company reports, industry statistics, trade journals, government publications, online databases.
Advantages	Tailored to the research question; up‑to‑date; control over quality.	Cheaper, quicker, often large sample sizes; useful for background information.
Limitations	Time‑consuming, costly, may suffer from non‑response bias.	May be outdated, not specific enough, or not fully reliable for the current market.
When to Use	When detailed, specific answers are required (e.g., testing a new concept).	When you need a broad overview or to benchmark (e.g., market size, competitor sales).

3.1 Checklist for Evaluating Secondary Sources

• Authority: Who produced the data? Is the source reputable?
• Timeliness: When was the data collected or published? Is it still relevant?
• Relevance: Does the data address the research question?
• Bias: Is there a commercial or political agenda that could colour the information?
• Methodology: Is the original data‑collection method described and appropriate?

4. Sampling

4.1 Why Sample?

Studying an entire population is usually impractical. A well‑designed sample provides a reliable snapshot of the whole, allowing conclusions to be generalised.

4.2 Sampling Frame

The sampling frame is the complete list of elements from which the sample is drawn (e.g., a customer database, electoral register). A clear frame reduces coverage error and improves representativeness.

4.3 Sampling Methods

Method	How It Works	Strengths	Weaknesses
Simple random sampling	Every individual has an equal chance of being selected.	Reduces selection bias.	Requires a complete sampling frame.
Systematic sampling	Select every kth person from an ordered list.	Easy to implement.	Can introduce periodic bias if the list has a pattern.
Stratified sampling	Divide the population into sub‑groups (strata) and sample from each.	Ensures representation of key segments.	More complex to design.
Convenience sampling	Select respondents who are easiest to reach.	Fast and inexpensive.	High risk of bias; limited generalisability.
Snowball sampling	Existing participants recruit further participants.	Useful for hard‑to‑reach groups.	May over‑represent similar viewpoints.

4.4 Sample Size Considerations

• Desired confidence level – usually 95 % (z‑value ≈ 1.96).
• Margin of error (confidence interval) – e.g., ±5 %.
• Population variability – more diverse populations need larger samples; expressed as the estimated proportion (p) in the formula.
• Practical constraints – budget, time, and accessibility.

For a proportion, a common approximation for required sample size (n) is:

\[ n = \frac{z^{2}\,p\,(1-p)}{e^{2}} \]

where z = z‑value for the confidence level, p = estimated proportion (use 0.5 for maximum variability), and e = desired margin of error.

5. Reliability & Validity of Data

• Reliability – the consistency or repeatability of a measurement. Example: If the same questionnaire is given to the same group on two occasions, similar answers indicate high reliability.
• Validity – the extent to which the instrument measures what it is intended to measure. Example: A question that asks “How often do you buy sports shoes?” is valid for measuring purchase frequency, but a leading question such as “Don’t you think sports shoes are essential?” reduces validity.
• Ways to check:
  • Pre‑test (pilot) the questionnaire to spot ambiguous wording.
  • Use established scales where possible (e.g., Likert statements that have been validated in previous research).
  • Triangulate with other data sources (e.g., compare survey responses with sales records).

6. Collecting Data – A Quick Overview

• Design the questionnaire or interview guide – use clear, unbiased wording; decide on closed vs. open questions.
• Pilot test – run a small trial to check for ambiguity, length and technical problems.
• Administer the instrument – online surveys, face‑to‑face interviews, telephone, mail, or observation.
• Record responses accurately – digital entry reduces transcription errors; back‑up data regularly.
• Check reliability and validity – repeatability, consistency checks and alignment with research objectives.

7. Analysing Quantitative Data

7.1 Common Statistical Tools

Tool	Purpose	Formula / Typical Output
Mean (Average)	Central tendency	\(\displaystyle \bar{x}= \frac{\sum_{i=1}^{n}x_i}{n}\)
Median	Middle value of ordered data	Position = \(\frac{n+1}{2}\) (odd n) or average of two middle values (even n)
Mode	Most frequent value	Value(s) with highest frequency
Range	Spread of data	Maximum – Minimum
Standard Deviation (σ)	Variability around the mean	\(\displaystyle \sigma = \sqrt{\frac{\sum_{i=1}^{n}(x_i-\bar{x})^{2}}{n}}\)
Percentage / Proportion	Part of a whole	\(\displaystyle \% = \frac{\text{Part}}{\text{Total}}\times100\)
Cross‑tabulation	Compare two categorical variables	Contingency table showing joint frequencies
Simple significance test (optional A‑level depth)	Assess whether an observed difference is likely due to chance	t‑test for means, χ² test for independence (compare calculated statistic with critical value at 5 % significance)

7.2 Step‑by‑Step Process

1. Data cleaning – check for missing values, outliers and entry errors; decide whether to correct, delete or retain them.
2. Organisation – enter data into a spreadsheet or statistical software; label variables clearly and code categorical responses.
3. Descriptive statistics – calculate mean, median, mode, range, standard deviation, percentages, and, where relevant, cross‑tabulations.
4. Visual presentation – use bar charts, histograms, pie charts, line graphs or scatter plots to highlight patterns.
5. Interpretation – identify trends, compare groups, and (if required) test significance (e.g., t‑test for two sample means, χ² test for association between two categorical variables).
6. Conclusions & recommendations – link the statistical findings back to the original research objectives and suggest actionable steps for the business.

8. Analysing Qualitative Data

8.1 Common Techniques

• Thematic analysis – coding responses and grouping similar ideas into themes.
• Content analysis – counting the frequency of specific words, phrases or concepts.
• Sentiment analysis – categorising statements as positive, negative or neutral.
• SWOT extraction – pulling out strengths, weaknesses, opportunities and threats from open‑ended feedback.

8.2 Step‑by‑Step Process

1. Transcribe interviews, focus‑group discussions or open‑ended questionnaire answers verbatim.
2. Read through the material several times to become familiar with the content.
3. Develop a coding framework – assign short labels (codes) to recurring ideas.
4. Apply the codes to each piece of data; remain flexible for new codes that emerge.
5. Cluster related codes into broader themes (and sub‑themes where useful).
6. Summarise each theme, selecting illustrative quotations that capture the essence of the group.
7. Interpret the themes in relation to the research objectives and the business context.

9. Integrating Quantitative and Qualitative Findings

• Quantitative results answer “what” is happening (e.g., 62 % prefer product A).
• Qualitative insights explain “why” it is happening (e.g., customers cite design and price).
• Triangulation – cross‑checking both data sets increases reliability and provides a richer, more nuanced picture.

10. Example: Analysing a New Smartwatch Survey

A company surveyed 200 potential customers about a forthcoming smartwatch.

10.1 Quantitative Summary

Question	Response option	Number of responses	Percentage
Would you purchase the smartwatch?	Yes	124	62 %
	No	76	38 %
Preferred price range	$150‑$200	85	42.5 %
	$200‑$250	70	35 %
	Above $250	45	22.5 %

10.2 Qualitative Themes (open‑ended question: “What would make you more likely to buy?”)

• Battery life – “If it lasts at least two days, I’d consider it.”
• Design aesthetics – “I want a sleek, minimalist look.”
• Health‑tracking features – “Accurate heart‑rate monitoring is essential.”
• Price sensitivity – “A discount or bundle would tip the balance.”

10.3 Integrated Interpretation

While a clear majority (62 %) express willingness to buy, the qualitative themes reveal two decisive barriers: battery life and price. The business should therefore:

1. Invest in battery optimisation to achieve at least a 48‑hour claim.
2. Consider promotional pricing or bundled accessories to attract the price‑sensitive segment (≈ 23 % who prefer > $250).
3. Highlight health‑tracking capabilities and design in marketing communications to reinforce the positive themes.

11. Common Pitfalls & How to Avoid Them

• Over‑reliance on the mean – check for skewed distributions; use median or mode where appropriate.
• Ignoring outliers – determine whether they are data‑entry errors or genuine insights before deciding to retain or remove them.
• Selecting only favourable quotations – ensure qualitative excerpts represent the full range of opinions.
• Confirmation bias – stay open to findings that contradict expectations; let the data speak.
• Insufficient sample size – a too‑small sample reduces confidence in the results; calculate the required size beforehand using the confidence‑level formula.
• Poor secondary‑source evaluation – use the checklist in 3.1 to guard against outdated or biased information.

12. Summary Checklist for Data Analysis

1. Validate and clean raw data (remove errors, decide on outliers).
2. Confirm that the sampling frame and method are appropriate; ensure the sample size meets the desired confidence level and margin of error.
3. Check reliability (repeatability) and validity (measuring the intended construct) of the instrument.
4. Choose suitable statistical measures for quantitative data; run simple significance tests where required.
5. Present quantitative results with clear tables and visualisations.
6. Code qualitative data systematically and develop coherent themes.
7. Use illustrative quotations that reflect the breadth of opinion.
8. Triangulate quantitative trends with qualitative explanations.
9. Draw concise, actionable recommendations linked directly to the research objectives.