A Marketing Management thesis needs variables built on a named, published marketing model — not an intuitive list of factors — because a model gives the instrument existing validated items and gives the discussion chapter an established body of findings to compare against. Below are three complete variable-and-hypothesis sets, each built on a different marketing theory, so you can see how a theory converts into variables, hypotheses and a statistical test, then adapt the closest one to your own topic.
How is this different from the site’s general BSBA hypotheses guide?
The site’s existing BSBA thesis hypotheses guide gives one worked example each across marketing, HR, finance and operations management — useful as a first orientation across the whole BSBA program, but necessarily shallow on any one area. This piece stays inside marketing only, with three full sets instead of one, each anchored to a specific, named model rather than a generic “hypotheses about marketing” framing, and adds the theory-selection and instrument-adaptation guidance the shorter general piece has no room for.
Set 1: Brand equity and purchase intention (Aaker’s brand equity model)
David Aaker’s brand equity framework names five asset categories that give a brand its value: brand awareness, brand loyalty, perceived quality, brand associations and other proprietary brand assets (patents, trademarks, channel relationships). Consumer-survey theses usually measure only the first four, since the fifth is not something a consumer can rate. A thesis studying, for example, a local food or beverage brand’s competitiveness against imported alternatives can operationalize three of Aaker’s dimensions as independent variables and purchase intention as the dependent variable. The respondent group this set implies is a brand’s actual or potential consumer base — typically intercepted in a retail or mall setting, or reached through an online consumer panel if the brand sells primarily through e-commerce.
| Variable | Type | Sample indicator |
|---|---|---|
| Brand awareness | IV | Respondent’s unaided and aided recall of the brand |
| Perceived quality | IV | Rated agreement on product consistency and reliability |
| Brand loyalty | IV | Repeat-purchase frequency and stated willingness to repurchase |
| Purchase intention | DV | Stated likelihood of purchasing in the next buying occasion |
Null hypotheses: H₀1 — Brand awareness has no significant relationship with purchase intention. H₀2 — Perceived quality has no significant relationship with purchase intention. H₀3 — Brand loyalty has no significant relationship with purchase intention. Statistical tool: Pearson’s r for each bivariate relationship (after a normality test), or multiple regression if the study also asks which of the three dimensions predicts purchase intention most strongly.
Set 2: Technology acceptance and digital marketing adoption (TAM)
The Technology Acceptance Model (Davis, 1989) is the standard framework for a thesis studying why consumers or small businesses adopt a digital marketing tool or platform — a social commerce feature, a QR-code payment-linked promotion, or a small business’s shift to social media selling. TAM’s two core constructs are perceived usefulness and perceived ease of use, both predicting behavioral intention to use. This set fits a respondent group of platform users or small business owners who have already tried the tool under study, since the constructs ask about experienced use rather than a hypothetical.
| Variable | Type | Sample indicator |
|---|---|---|
| Perceived usefulness | IV | Rated agreement that the platform improves sales or reach |
| Perceived ease of use | IV | Rated agreement that the platform is simple to learn and operate |
| Behavioral intention to use | DV | Stated intention to continue or expand use of the platform |
Null hypotheses: H₀1 — Perceived usefulness has no significant relationship with behavioral intention to use. H₀2 — Perceived ease of use has no significant relationship with behavioral intention to use. Statistical tool: Pearson’s r or Spearman’s rho depending on the normality test result; a small-business sample studying actual adoption behavior over time could instead use a chi-square test of independence between ease-of-use rating (categorized) and adoption status (adopted/not adopted).
Set 3: Service quality and customer satisfaction (SERVQUAL / the extended marketing mix)
For a thesis on a service business — a café, a salon, a co-working space — the SERVQUAL dimensions (tangibles, reliability, responsiveness, assurance, empathy) already used on this site for a tourism thesis translate directly to a marketing-management study of customer satisfaction and repeat patronage, and can be paired with the extended marketing mix’s People, Process and Physical evidence elements where the study is explicitly framed around the 7Ps rather than SERVQUAL alone. The respondent group here is the business’s own actual customers, typically intercepted on-site immediately after a service encounter, while the experience is still fresh enough to rate accurately.
| Variable | Type | Sample indicator |
|---|---|---|
| Service quality (SERVQUAL composite or a single dimension, e.g. responsiveness) | IV | Rated agreement on staff responsiveness to requests and complaints |
| Price perception | IV | Rated agreement that pricing is fair relative to the service received |
| Customer satisfaction | DV | Overall rated satisfaction and likelihood to recommend |
Null hypotheses: H₀1 — Service quality has no significant relationship with customer satisfaction. H₀2 — Price perception has no significant relationship with customer satisfaction. Statistical tool: Pearson’s r or Spearman’s rho for each relationship, or multiple regression where both predictors are tested together against satisfaction.
Writing the hypothesis sentence itself
A null hypothesis states no significant relationship or difference; the alternative hypothesis, which is what the study actually expects to find support for based on the literature, states that a relationship or difference does exist. Philippine thesis manuscripts conventionally report the null form in the hypotheses section of Chapter 1, since that is the form the statistical test technically evaluates — the test either rejects or fails to reject the null, it does not directly “prove” the alternative. Keep the hypothesis worded at the same level of specificity as the variable itself: “brand awareness has no significant relationship with purchase intention” is testable; “branding affects buying” is not specific enough for a panel to know which statistical test answers it. State the significance level the test will be evaluated against — conventionally α = 0.05 for an undergraduate marketing thesis — once in Chapter 3, rather than repeating it after every individual hypothesis.
Adapting the instrument: why a named model still needs local validation
Borrowing a published model’s items is the right starting point, but a scale developed and validated abroad — most TAM and SERVQUAL instruments were first validated on non-Philippine samples — still needs the same content-validation and pilot-testing pass any adapted instrument needs: at least three expert validators checking the items translate cleanly and mean the same thing to a Filipino respondent, a pilot test on a small group outside the final sample, and a reported Cronbach’s alpha per construct rather than one pooled figure across every variable in the study. A discussion chapter that compares your own alpha and mean scores against the original study’s published figures, rather than treating the borrowed scale as automatically valid in a new context, is what a statistics-literate panel is actually checking for. Where an item genuinely does not translate — a Western retail-context question that assumes a shopping mall format not present in a rural municipality, for instance — note the adaptation explicitly in Chapter 3 rather than forcing the original wording to fit a setting it was not written for.
Choosing between the three sets, or building your own
Match the model to the business context you can actually access data from: brand equity fits a study comparing consumer perception of competing products or brands, TAM fits any study about adopting a new marketing technology or channel, and SERVQUAL or the 7Ps fits a service business where the “product” is an experience rather than a physical good. Where none of the three fits cleanly, other established models remain available — the AIDA model (Attention, Interest, Desire, Action) for advertising-effectiveness studies, or the Theory of Planned Behavior for studies of purchase intention driven by attitude, subjective norm and perceived behavioral control — but whichever is chosen, name the model, its original author and year in the theoretical framework, the same standard the site’s conceptual framework guide holds every discipline to. A model chosen because it is genuinely the closest fit to the business context, rather than because it is the most familiar one from a marketing textbook, is what a panel reads as a deliberate methodological choice rather than a default.
From hypotheses to the rest of Chapter 1 and Chapter 3
Once the variable set is chosen, the statement of the problem’s specific questions should mirror the hypotheses one for one, following the same worked pattern in the site’s statement-of-the-problem guide, and the operational definitions and instrument items belong in a consistency matrix built the way the site’s operationalization of variables guide demonstrates. For a title bank across marketing and other BSBA areas with feasibility notes already worked through, the site’s BSBA and HRM title examples guide covers forty additional starting points.
Build your own variable set from your own topic
Tesify drafts the hypotheses, conceptual framework and Chapter 3 instrument sections from your own marketing topic, matched to the theory that actually fits your business context. Over 9,000 students have used it to write more than 15,000 chapters, and every word is still written by you.
Frequently asked questions
How is this different from the site’s general BSBA hypotheses guide?
The general BSBA guide gives one worked example each across marketing, HR, finance and operations — a shallow pass across four areas. This piece goes deep on marketing specifically, with three full variable-and-hypothesis sets built on named marketing theories, which the general guide does not have room to cover.
Do I need to use a named theory like TAM or Aaker’s model, or can I build my own variables?
A named, published model is strongly preferred over a self-built variable list, because it gives your instrument existing validated items to adapt and gives your discussion chapter an established body of findings to compare against; a self-built list is harder to defend when a panel asks where the variables came from.
Can I combine two theories, like TAM and brand equity, in one thesis?
Yes, but only where the combination is logically justified — for example, brand equity influencing the perceived usefulness of a branded app — and the conceptual framework must show how the two models connect rather than listing them side by side with no stated relationship.
What is the most common hypothesis-writing mistake in a marketing thesis?
Writing a hypothesis about a relationship the chosen statistical test cannot actually establish — claiming one variable “causes” another from a correlational, non-experimental survey design, when the test only supports a claim of association.
Do I need to adapt a foreign-validated scale before using it on a Filipino sample?
Yes. Re-run content validation and a pilot test even for a well-established scale like TAM or SERVQUAL, since a scale validated abroad is not automatically valid for a new cultural or linguistic context, and report your own reliability figures rather than citing only the original study’s.