Mastering your quantitative research thesis from A to Z

Start with a broad topic in your field, like executive management, then narrow to a specific, achievable topic while considering scope, interest, access, and ethics to shape the research question.

Identify a topic by weighing relevance, interest, specificity, and ethical grounding; brainstorm broadly, assess trends, and discuss ideas with peers, practitioners, and supervisors to narrow to a specific, achievable direction.

Identify a precise research question and uncover a research gap by reviewing literature, then specify concepts like idea generation and leadership to frame your thesis.

Identify the flexible parts of your research question to adjust for data, access, and literature. This flexibility guides sampling and context choices, such as multinational teams or German companies.

Perform a scoping literature review to align your research question with academic language and debates. Rely on international peer-reviewed journal publications as main source to validate concepts and terms.

Learn to use a literature management software to streamline your thesis reference process and reduce errors, with open source options like Sotero and easy-to-use Jiru.

Draft the introduction and theoretical background by outlining the research question, gap, and topic relevance, then define dependent and independent variables and propose hypotheses.

Starting to write now clarifies and restructures complex research ideas, because writing is a form of thinking that helps you think more clearly, beyond simply disseminating results.

Identify data collection options for addressing a research question by using existing data through secondary data analysis or by collecting new primary data, with both approaches explored.

Explore how measurement frequency shapes research design by contrasting longitudinal and cross-sectional approaches, weighing causality, multiple time points, data collection, and privacy logistics.

Compare surveys and observations as data collection methods in quantitative research, and decide when to observe versus ask about concepts like attitudes and opinions, and ideal measurements for your concepts.

Learn to define your population, select a sampling frame, compare sampling methods (random, quota, snowball, and convenience), and set a justified sample size for robust thesis research.

Apply power analysis and a one-to-20 (or one-to-10) cases rule of thumb to determine sample size in quantitative analysis, accounting for population diversity, effect size, and model complexity.

Design measurement instruments by using simple questions for easy counts, or multi-item scales for complex concepts, treating observation sheets like surveys and averaging items for robust scores.

Reuse established measurement instruments to align with existing science, saving time and supporting cumulative research; adapt them to your population and rigorously test them with feedback.

Apply cognitive pretest using think-aloud protocol to observe how respondents interpret survey concepts, note comprehension problems, and revise the measurement instrument through multiple iterations to boost data quality.

Develop your methods section by detailing data collection and sampling with sample demographics, describe measurement instruments and item development, and outline data analysis procedures.

Discover JASP, a free statistical package that uses a user interface built on R, delivering robust calculations with continuous updates.

Download JASP for Linux, Mac OS, or Windows, then install it, or run JASP online via the free Rawle app registration, and launch online to try it.

Explore the graphical user interface of the statistics software, load data, manage datasets, and enable modules like summary statistics or structural equation modeling for basic statistics and classical testing theory.

Navigate the four preference blocks to manage data synchronization, note the limited spreadsheet editor, fix decimals to three, and set language and theme for your workflow.

Navigate to the main menu, open data from your computer, and load your dataset, while exploring recent files and open science framework integration for project data and sample datasets.

Explore the data set structure by reading rows as respondents and columns as variables, apply filters, and understand nominal, ordinal, and scale variable types.

Explore descriptive statistics basics by describing data with minimal figures, monitor real-time outputs for means and standard deviations, and review frequency tables and data grouping.

Explore descriptive statistics through plots such as box plots, scatter plots, and distribution plots, noting medians, quartiles, interquartile ranges, outliers, and potential gender differences.

Identify the key variables, use box plots to detect outliers, and assess the distribution with standard error of the mean and standard deviation.

Perform a one-sample t-test to compare sample means against a test value, interpret p-values and test statistics, and use descriptives, normality checks, and the Wilcoxon test when needed.

Perform an independent samples t-test to compare means between two groups using a grouping variable, verify normality and equal variances, and note non-parametric or Welch alternatives when assumptions fail.

Learn how to perform a t-test for dependent (paired) samples, interpret t and p values, and use a Wilcoxon signed-rank test as a nonparametric alternative with effect size.

Explore analysis of variance (ANOVA) to compare multiple groups, interpret the F value and p value, and apply Tukey post hoc corrections to control alpha inflation while checking assumptions.

Explore multiple linear regression to predict objective job performance from learning behaviors, using continuous and binary predictors, and assess model fit, coefficients, and key assumptions.

Learn how to conduct mediation analyses in chaske using structural equation modeling to explain how gender affects objective job performance through learning behaviors, with direct, indirect, and total effects.

Finish your methods section by outlining sampling procedure, measurement instruments, and the analysis, focusing on standard statistical methods like correlations and regressions, and showing how they answer your research question.

present your data in the results section using means, standard deviations, regression coefficients, and p values, describe them verbally, and state whether hypotheses were accepted or not; keep it concise.

Uncover three views on mixing methods in statistical software like JASP and SPSS: yes, no, or maybe within mixed methods designs that integrate quantitative and qualitative data.