From three questions to a model that can be tested.
A research question becomes testable once it is drawn as a model: which things are expected to influence which, in what order, and why. This page shows the framework I chose, the model I built on it, and the reasoning behind each of its thirteen paths.
The framework: stimulus, organism, response
The study uses the stimulus, organism, response (S-O-R) framework from environmental psychology, which is widely used in retail and online shopping research. Its logic is simple: features of an environment trigger internal states in a person, and those internal states produce behaviour. It fits this study because it keeps three things apart that are easy to blur: what the app offers, what happens inside the shopper, and what the shopper then intends to do.
What the environment offers
Cues that the person encounters and evaluates. In an online setting these are measured as the shopper perceives them.
In this study: useful, enjoyable and social value of mobile commerce.
What happens inside the person
The thinking and feeling that the cues set off. This is where the framework allows both cognitive and emotional reactions.
In this study: immersion, searching to decide, browsing to discover.
What the person does next
The approach or avoidance behaviour that follows from those internal states.
In this study: intent to return and recommend.
Select an arrow label or line to read the hypothesis behind it.
The hypotheses and the reasoning behind them
Thirteen direct paths and six mediation tests. Each group below gives the formal statements and, in a few sentences, the argument that the thesis develops in full in Chapter 3.
A useful app invites exploration and immersion
- H1aFunctional value increases searching to decide.
- H1bFunctional value increases browsing to discover.
- H1cFunctional value increases immersion.
Why I expected this. Access anywhere removes the limits of time and place, personalization shortens the route to relevant content, and clear navigation cuts the effort of finding things. Less effort means more exploring, and a smooth, responsive experience gives the sense of control that immersion depends on.
An enjoyable app invites exploration and immersion
- H2aEmotional value increases searching to decide.
- H2bEmotional value increases browsing to discover.
- H2cEmotional value increases immersion.
Why I expected this. When browsing is fun, it becomes its own reward, so people do more of it. Visual appeal creates the first impression that pulls a shopper in, and rich product presentation helps them evaluate detail. Enjoyment and captured attention are also two of the classic preconditions of flow.
A social app invites exploration and immersion
- H3aSocial value increases searching to decide.
- H3bSocial value increases browsing to discover.
- H3cSocial value increases immersion.
Why I expected this. Reviews, ratings and quick two-way communication are sources of information that fill gaps in what the shopper knows. A sense of human warmth builds trust and keeps people in the environment longer, and interactivity adds to the feeling of control that supports flow.
Immersion leads to both kinds of exploration
- H4aImmersion increases searching to decide.
- H4bImmersion increases browsing to discover.
Why I expected this. Flow combines enjoyment with deep concentration. The enjoyment side counters boredom, which is what drives discovery browsing. The focus and control side counters overload and anxiety, which is what drives focused searching. One state should therefore feed both behaviours.
Immersion carries the effect of value
- H5Immersion mediates the effect of each kind of value on searching to decide.
- H6Immersion mediates the effect of each kind of value on browsing to discover.
Why I expected this. The original model of online flow places it between the features of an environment and exploratory behaviour. If that holds on mobile, value should work at least partly by first drawing the shopper into flow. That gives six mediation tests: three kinds of value by two kinds of exploration.
Both kinds of exploration lead to return and recommend
- H7Searching to decide (specific exploration) increases future patronage intention.
- H8Browsing to discover (diversive exploration) increases future patronage intention.
Why I expected this. Focused searching resolves uncertainty: shoppers who find the answer they needed grow familiar with the platform and confident in their decision. Discovery browsing is enjoyable in itself and builds a mental catalogue of what is available. Both are satisfying experiences, and satisfying experiences are the ones people repeat and recommend.
Four modelling decisions worth explaining
Perceptions as the stimulus
The study measures value as the shopper perceives it, not features that were manipulated in a lab. That is the standard survey-based use of S-O-R, and it suits a question about mobile commerce in general and not about one particular app.
Exploration sits in the organism
The two kinds of exploration are treated as internal states, one mainly cognitive (searching to decide) and one mainly affective (browsing to discover). The behavioural response is what comes after: the intention to return and recommend.
Value as three higher-order constructs
Each kind of value is built from distinct dimensions that are not interchangeable: for example, usefulness from ubiquity, customization and navigation. Modelling them as formative higher-order constructs keeps the model compact while showing which dimension carries the weight.
Flow as one construct, not a few picked dimensions
Earlier studies measured flow with two or three chosen dimensions and got inconsistent results. Here flow is a single construct formed by eight facets, from control and focus to the sense that time is passing quickly.
Six control variables were included so that the paths are not simply reflecting who the respondent is: age, gender, education, mobile internet plan, individualism-collectivism and uncertainty avoidance.
What the model is built on
| Building block | Core idea taken forward | Key sources |
|---|---|---|
| S-O-R framework | Environmental cues shape internal states, which shape approach behaviour; adapted to online stores. | Mehrabian and Russell (1974); Eroglu, Machleit and Davis (2001) |
| Perceived value | Value is multidimensional: functional, emotional and social. | Sheth, Newman and Gross (1991); Sweeney and Soutar (2001) |
| Flow | Optimal experience, and its role in computer-mediated environments. | Csikszentmihalyi (1990); Hoffman and Novak (1996) |
| Exploration | Diversive and specific exploration as two responses to stimulation; applied to online shopping. | Berlyne (1960); Demangeot and Broderick (2010) |
| Patronage intention | Willingness to return, buy and recommend as a single outcome. | Baker, Parasuraman, Grewal and Voss (2002) |