Quantitative and Qualitative Analysis

Data analysis is another core component of a dissertation, whereby raw data is cleansed, coded, processed and changed to an interpretable form, known as information. Depending on the nature of the data, i.e., quantitative (numerical) or qualitative (non-numerical), a wide variety of analysis techniques may be applied.

It is important for students to understand that, despite the multitude of YouTube video tutorials on statistical analysis techniques, what matters at the end is the proper and clear interpretation of results. This is not within the grasp of just anyone. Therefore, when in doubt, contact us at Stat Consul.

Quantitative Analysis

Quantitative data is numerical in nature, as a result of the coding of primary data retrieved from survey questionnaires or secondary data downloaded from websites. These are analysed by software like Excel, SPSS, Stata, SAS and R, to name a few. At Stat Consul, we analyse numerical data using SPSS and Excel.

Descriptive Analysis

Descriptive analysis, in its basic form, entails summarising responses as tables of frequencies and/or percentages, illustrating them on charts or calculating summary statistics (mean, standard deviation, skewness, kurtosis, etc.). In the case of sets of Likert-type statements, the most effective descriptive analysis technique is the method of weighted means (Glen, 2014), whereby statements are ranked according to their weighted means, prior to interpretation.

Inferential Analysis

Inferential analysis used to generalise the results obtained from a probability sample back to its parent population. This type of analysis is more powerful than its descriptive counterpart, as results are more evidence-based and conclusive. Very often, through hypothesis testing, inferential techniques help establish relationships among research constructs and also make predictions.

There are numerous inferential techniques that may be used for your dissertation, depending on your research objectives, as well as on how your measuring instruments are designed:

• Parametric tests: paired t-test, independent-samples t-test, one-way ANOVA
• Non-parametric tests: Chi-Squared test of independence, Mann-Whitney U test, Kruskal-Wallis H test, Wilcoxon Signed Rank Test
• Multivariate techniques: Correlation analysis, two-way ANOVA, MANOVA, exploratory factor analysis (EFA), multiple regression analysis (OLS), logistic regression, structural equation modelling (SEM)

Note
At Stat Consul, we always test your data for internal consistency, construct validity, sample adequacy and normality before conducting quantitative analysis.

Qualitative Analysis

Qualitative data may be collected by observation or by means of in-depth interviews, focus group discussions, open-ended surveys and case studies. In all, if not most, cases, collected data is in the form of texts or transcripts. Data may then be analysed by way of thematic analysis (Braun and Clarke, 2006) or by software like NVivo, which will reorganise unstructured text, audio, video, and image data, including (but not limited to) interviews, focus groups, surveys, social media, and journal articles.