Comparing qualitative and quantitative CR&S data

Quantitative and qualitative data answer different questions and comparing them well begins with understanding what each does. Quantitative data is numerical and measurable. It shows scale, trend and comparison, can be aggregated, benchmarked and tracked over time. Its limitation is that it rarely explains itself, a fall in employee turnover is a number, but it does not say why people stayed. Qualitative data is descriptive: it captures experience, motivation, context and meaning. Its limitation is that it is harder to compare, can be subjective, and cannot easily be scaled or verified. Because CR&S performance involves both measurable impacts and human experience, neither type alone is sufficient, the practitioner’s task is to compare and combine them (Jick, 1979).

The foundational technique for comparing the two is triangulation: using multiple data sources or methods to examine the same question and seeing whether they point to the same conclusion (Jick, 1979; Scribbr, 2023; Simply Psychology, 2024). If emissions data, an energy audit and staff accounts of building conditions all point to the same problem, the conclusion is strong because it is corroborated from different directions. Triangulation increases confidence in findings by cross-verifying them (Scribbr, 2023). It is the single most important principle in comparing CR&S data, and it works precisely because it draws on evidence of different kinds.

A second technique is to convert qualitative data into a form that can be compared with quantitative data. Thematic analysis (also called coding) is the systematic method for doing this: qualitative material such as survey comments, interview notes or community feedback is read closely, recurring themes are identified and labelled, and the frequency or strength of each theme is recorded (Braun and Clarke, 2006; ATLAS.ti, 2024). This turns unstructured text into structured evidence that can be counted, compared and tracked. For example, coding 200 employee comments might reveal that 40% raise concerns about workload, a finding that can then be set alongside quantitative wellbeing or absence data (ATLAS.ti, 2024). Sentiment analysis and simple scoring scales perform a similar function, allowing qualitative input to be compared numerically.

A third technique is the structured comparison of findings to look for convergence and divergence. Convergence is where qualitative and quantitative data agree and reinforce each other, this strengthens a conclusion. Divergence is where they disagree and this is often where the most valuable insight lies (Jick, 1979). If quantitative data shows a diversity policy is being met on paper but qualitative feedback reveals employees do not feel included, the contradiction itself is the finding, it tells the practitioner that the metric is not capturing the real situation. A practitioner who treats divergence as a signal to investigate further, rather than an inconvenience to explain away, produces far more credible conclusions.

Several further techniques support comparison. Data visualisation such as charts, dashboards and thematic maps allows quantitative trends and qualitative themes to be presented side by side so patterns become visible. Weighting and materiality help decide how much importance to give each piece of evidence. A serious qualitative concern raised by a small number of people may matter more than a marginal movement in a metric, depending on the issue’s materiality. Benchmarking sets quantitative performance against peers or standards, while qualitative context explains why an organisation sits where it does. Across all of these, the principle is the same: the two data types are not rivals to be ranked, but complementary evidence to be combined.

Forming evidence-based conclusions is the purpose of the whole exercise. A sound conclusion is one that is supported by more than one type of evidence, that acknowledges where evidence conflicts, and that is proportionate to the strength of the data behind it. The most common failures are selecting only the data that supports a desired conclusion and stating a conclusion more confidently than the evidence allows. A credible CR&S practitioner draws conclusions honestly: stating what the combined evidence shows, how confident they are in it, where the data agrees and disagrees, and what further evidence would be needed to be more certain. This disciplined comparison of qualitative and quantitative data is what separates genuine insight about CR&S performance from selective storytelling, and it is the analytical foundation for the reporting activities that follow in this learning cycle.