Undergraduate Thesis: Structure, Writing Guide, and Detailed Example

What is an undergraduate thesis in a dissertation?

An undergraduate thesis is a research project completed during the final stage of an undergraduate program. You use subject knowledge and research methods to answer a specific question, based on literature, data, and an analysis process that can be checked. For a quantitative topic, the thesis usually includes a questionnaire, .sav or .csv data, results from SPSS or SmartPLS, and chapters formatted according to your department's requirements.

Many students search for “what is an undergraduate thesis” because they are deciding between a thesis and an internship report. An internship report mainly describes the host organization, the work completed, and the experience gained. An undergraduate thesis focuses on the research problem, model, hypotheses, scales, data, and evidence-based conclusions. Some programs combine an internship with a thesis, but you still need to check the proposal template and the specific rules for your program.

A quantitative thesis is usually longer than an internship report because it must explain both the theoretical foundation and the analysis procedure. The exact length depends on your university, field, and supervisor. Check your department's formatting guide before finalizing the page count, number of chapters, citation style, and appendices.

If you are unsure where to begin, review how to identify a research topic and the list of scientific research topics to check whether the problem, available data, and remaining time fit together. A topic may sound interesting but still create problems in every later chapter if you cannot collect enough data.

Standard structure of an undergraduate thesis

A quantitative thesis usually follows a logic that moves from the problem to evidence and then to the conclusion. Chapter titles may differ across universities, but the core content is usually similar.

The table is a working framework, not a mandatory page requirement. If your university requires five chapters, the theoretical foundation and literature review may be separated. If the results and discussion must be combined, you should still distinguish the paragraph that reports the result from the paragraph that explains its meaning.

You should also distinguish an undergraduate thesis from a master's thesis. Both require an argument and evidence, but the educational level, methodological requirements, theoretical depth, and formatting rules may differ. Do not copy the structure of a graduate thesis if your department has a separate template for an undergraduate thesis.

Steps for writing an undergraduate thesis

Step 1: Finalize the problem, objectives, and research questions

Write a short paragraph that answers three questions: what phenomenon are you studying, what do you want to explain, and who will use the result. Then turn the general objective into two to four specific objectives. Each objective should be testable with the data you can actually collect.

For example, the objective “to assess the factors affecting university students' intention to use e-wallets in Ho Chi Minh City” can be developed into smaller objectives: identifying the factors, measuring the strength of their effects, testing model fit, and proposing implications. This gives the results chapter a clear point of reference instead of leaving you with a series of tables and no way to show which objective each table addresses.

Step 2: Find literature and build the model

Create a tracking table with the study title, context, variables, scales, method, and main findings. Keep sources that directly support the concept or relationship you intend to test. A useful literature review is not a list that summarizes one paper after another. It shows the research gap and explains why your model is reasonable.

When you use online sources, save the bibliographic information as soon as you read them. The in-text citations must match the reference list. You can review the guide to checking Turnitin when screening for overlap, but a similarity score is only a signal that needs review. You still need to check your interpretation, citations, and the sections you wrote yourself.

Step 3: Design the scales and questionnaire

In quantitative research, each concept must be converted into measurable items. Identify the source of each scale, translate and adjust the wording for the Vietnamese context, then design the questionnaire with screening questions, demographic information, and measurement statements. A five-point or seven-point Likert scale should be explained consistently at the beginning of the measurement section, using a design appropriate to the topic.

At this stage, state which variables are independent, dependent, mediating, or moderating variables. If an item is reverse-coded, mark it in the data file and process it before running Cronbach's Alpha. A questionnaire that is too long can cause respondents to abandon it, while one that is too short may fail to cover the concept adequately.

Step 4: Collect and clean the data

Set your validity criteria before collecting responses. For example, respondents may need to belong to the correct target group, complete the main section, and avoid submitting more than once. After exporting the data, check rows with extensive missing values, identical answers across every item, unusual completion times, and incorrect coding.

If the questionnaire is distributed online, fillform.info is DoThesis's form tool for collecting responses in Vietnamese. Google Forms is a familiar alternative. Whichever tool you use, save the original data file, the cleaned data file, and a log recording every change.

Step 5: Run the analysis in the correct order

With SPSS, the process usually moves from descriptive statistics to reliability testing, EFA if required by the model, and then correlation and regression. With SmartPLS, assess the measurement model first and then the structural model. This two-step approach follows the logic of assessing measurement before the structural model (Anderson and Gerbing, 1988).

Do not delete an item simply because you want a cleaner table. Every deleted item needs a reason based on the statistic, the content of the scale, and its fit with the model. Save the output from every run and use clear names such as alpha_v1, efa_v2, or pls_final, so you can explain what changed when your supervisor asks.

Step 6: Write results linked to the objectives

Every table in the results chapter needs a number, title, source or note, and an explanation immediately after it. The explanation should state what the table shows, which variables meet or fail the criterion, and which hypothesis the result relates to.

You can present a shortened output table, but do not paste an entire SPSS screen into the main text. Put the complete output in the appendix, while the results chapter retains the columns and figures needed for the argument.

Step 7: Discuss the results, implications, and limitations

The discussion should not repeat the results word for word. Explain why a relationship may be significant, compare it with previous research, and identify similarities or differences in your context. Managerial implications must relate to the variables with effects and to a specific target group. Avoid general statements such as “businesses need to improve quality”.

The limitations section should state what the data cannot answer, such as convenience sampling, a narrow geographic scope, a cross-sectional design, or a limited number of variables. Stating a limitation does not weaken the thesis when you explain how it affects generalizability.

Reusable sentence templates

The following sentences are frameworks. Replace them with the actual information from your file. Do not enter figures before completing the analysis.

How to write this in your thesis

After replacing every section in square brackets, you can use the following structure: “The analysis shows that [independent variable] has a [positive or negative] effect on [dependent variable], with β = [value] and p-value = [value]. Therefore, hypothesis [H1] is [supported or rejected] in the research context of [target group and location].”

This sentence framework helps you keep the reporting logic clear. The figures, direction of the effect, and conclusion must come from the actual output. If the p-value does not meet the criterion specified in your method, report that the result is not statistically significant instead of changing the wording to support the hypothesis.

Short undergraduate thesis example

Suppose you are studying the factors affecting university students' intention to continue using e-wallets in Hanoi. The model includes perceived usefulness, perceived ease of use, trust, and intention to continue using. The questionnaire was distributed to students who had used an e-wallet, and the cleaned data were entered into SPSS.

The introduction presents the issue of e-wallet use and the gap concerning students in the research location. The theoretical foundation chapter explains the concepts, summarizes previous studies, and proposes hypotheses H1 to H3. The method chapter describes the sampling method, Likert scale, collection process, and analysis plan.

In the results chapter, first describe gender, age, usage frequency, and the type of wallet used. Next, report scale reliability, the EFA or measurement model results depending on the design, and then the model test. The table below is illustrative output, not the result of a real study.

Based on the illustrative table, the paragraph could read: “The regression results show that perceived usefulness has a positive effect on intention to continue using, with β = 0.31 and Sig. = 0.000. Because Sig. is below the selected significance level, H1 is supported in the research sample.” In the actual thesis, replace every figure with the number shown in your own Coefficients table.

The discussion explains why perceived usefulness has a higher coefficient than the other factors and compares the result with the literature presented in Chapter 1. The implications section may recommend improving functions, security information, or the user experience, but each recommendation must come from the result and relate to a specific target group.

Common errors that lead to committee questions

The first error is inconsistency between the topic title, objectives, model, and questionnaire. The title may refer to purchase intention while the questionnaire measures actual purchase behavior. The objective may say “identify the factors”, while the results only report descriptive statistics. Before submitting, create a comparison table linking the objectives, hypotheses, variables, analyses, and results tables.

The second error is presenting a threshold without its source or applying one threshold to every context. Cronbach's Alpha from 0.7 is generally accepted according to (Nunnally, 1978), while Alpha from 0.6 may be considered for a new scale or exploratory research according to (Hair et al., 2010). Corrected item-total correlation from 0.3 is supported by (Nunnally and Bernstein, 1994). Your stated criteria and conclusions must be consistent.

The third error is pasting output without explaining it. The committee may ask, “Which objective does this table answer?” or “Why was this item deleted?” Record the reason for every decision, including the deleted item, run number, and the statistic before and after processing.

The fourth error is confusing statistical significance with the size of an effect. A small p-value provides evidence of a relationship in the sample under the tested model, but it does not automatically mean that the effect is large or practically valuable. In the discussion, consider the coefficient, direction of the effect, and context together.

The fifth error is using unchecked sources, incomplete citations, or copied paragraphs from another thesis. You can review the content through plagiarism checking, but the final check must still include the reference list, table notes, and explanations written in your own words.

Another error is treating an internship report and an undergraduate thesis as the same document while ignoring departmental requirements. If you are unsure whether to write an undergraduate thesis, compare the assessment criteria, access to data, remaining time, and your supervisor's requirements. You can also review the Undergraduate Thesis topic collection before registering.

Frequently asked questions

What is an undergraduate thesis, and how is it different from an internship report?

An undergraduate thesis is a research project with a question, objectives, method, data, and conclusion. An internship report usually focuses on the host organization, completed work, and experience. A program may require both documents or allow the internship report to support the thesis, so check your department's rules.

Should I write an undergraduate thesis?

Consider your learning goals, available time, ability to collect data, and required output. A thesis may suit you if you want to practice research, need an academic product for further study, or can work consistently with your supervisor. If the data are not feasible or the deadline is too short, an alternative format allowed by your university may be safer.

How many chapters does an undergraduate thesis usually have?

Many quantitative theses use four or five chapters, including the introduction, theoretical foundation, method, results, discussion, and conclusion. The sections may be combined or separated depending on the department. Use the official proposal template instead of copying the number of chapters from a thesis in another field.

Should I use SPSS or SmartPLS for my undergraduate thesis?

SPSS is suitable for descriptive statistics, Cronbach's Alpha, EFA, correlation, regression, and several common tests. SmartPLS is suitable when the model includes latent variables, a measurement model, and a structural model based on PLS-SEM. Choose the tool according to the model, scales, and supervisor's requirements, rather than choosing it because one software produces better-looking tables.

What should I do if the data do not support a hypothesis?

Report the result exactly as it appears in the output, then discuss possible explanations such as the context, sample, scale, or research design. Do not delete data, change the hypothesis after seeing the result without recording the change, or edit the p-value. An unsupported hypothesis is still an interpretable result when the research process is transparent.

Open your proposal and data file now, create a table linking the objectives, hypotheses, variables, and results tables, and write each paragraph from the actual output. If you need help running the analysis on a .sav or .csv file, see DoThesis M4 data analysis.