A dissertation on job satisfaction is a common choice in MBA human resource management, and an easy one to get wrong, because “job satisfaction” is not one thing. It is an attitude with several facets, measured by named scales, framed by named theories, and it can be explained by a dozen different independent variables. This guide takes the construct apart so that you can write your Chapter 1, 2 and 3 around it: the definition, the dimensions, the theories your guide will expect, the scales you can actually use, hypotheses you can test and ten titles to adapt.
The examples are illustrative and name no real findings. Where a scale, theory or definition is attributed, it is attributed to the source it came from; where you must supply a detail, it is marked in square brackets as yours to supply.
What does “job satisfaction” mean in a dissertation?
The definition most often quoted in Indian dissertations is Edwin Locke’s (1976): job satisfaction is “a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences”. Two things in that sentence matter for your design. It is an emotional state, so you are measuring an attitude, not performance. And it results from an appraisal, so it depends on what the employee expects and values, not only on what the job provides.
That second point is why a study of two banks with the same salary scales can find different satisfaction. Locke’s range-of-affect idea, associated with the same 1976 chapter, treats satisfaction as the gap between what a person wants from a facet of the job and what they get, weighted by how much that facet matters to them. If you adopt it, your Chapter 2 should say so and your questionnaire should ask about importance as well as experience, or at least explain why it does not.
Whichever definition you use, write it as a conceptual definition with a citation, then write your operational definition separately: which scale, which score, which cut-off. The distinction is worked through, with examples from other fields, in our guide to operational definitions for an Indian thesis.
The dimensions you can measure
Most dissertations treat job satisfaction in one of two ways: a single overall score, or a set of facets. Facets are usually the better choice for a dissertation, because they give you something to compare and to explain. The facet structure you adopt should come from a named instrument, not from a list you assembled. Three are widely cited:
| Instrument | Authors and year | What it measures | Format |
|---|---|---|---|
| Job Satisfaction Survey (JSS) | Spector (1985), American Journal of Community Psychology, 13, 693–713 | Nine facets: pay, promotion, supervision, fringe benefits, contingent rewards, operating procedures, coworkers, nature of work, communication | 36 items, four per facet, six response choices from strongly disagree to strongly agree; about half the items reverse-scored |
| Job Descriptive Index (JDI) | Smith, Kendall and Hulin (1969) | Five facets: pay, promotions, coworkers, supervision, the work itself | Yes, no or uncertain responses to descriptive statements |
| Minnesota Satisfaction Questionnaire (MSQ) | Weiss, Dawis and England (1967) | Twenty facets plus intrinsic and extrinsic satisfaction subscales | Long and short forms |
The JSS is the one many MBA scholars reach for first. Its author states on his own site that he permits use in return for sharing of results, appropriate attribution through citation and inclusion of the copyright notice, and that the survey exists in more than thirty languages. Read those conditions on the author’s page before you rely on them, and keep the copyright notice on the instrument you put in your appendix. For the JDI and the MSQ, confirm the publisher’s permission terms before you adopt them, because both are published instruments with their own conditions.
If none of them fits your sector, the alternative is to build your own tool from the literature, which is a legitimate route but a longer one. The steps, from blueprint to try-out, are in our guide to tool construction and validation for an Indian dissertation.
Which theories frame job satisfaction?
An examiner reading “theoretical framework” in an MBA dissertation wants two or three theories that explain why satisfaction rises or falls, each linked to a variable in your model. Three are standard:
- Herzberg’s motivator-hygiene (two-factor) theory, from Herzberg, Mausner and Snyderman’s The Motivation to Work (1959). It separates motivating factors, which are intrinsic, such as achievement and recognition, from hygiene factors, which are extrinsic, such as pay and working conditions. Use it when your independent variables fall naturally into intrinsic and extrinsic groups.
- Locke’s range-of-affect (discrepancy) theory (1976). Satisfaction depends on the gap between what is wanted and what is received on each facet, weighted by importance. Use it when you want to explain why the same condition satisfies one group and not another.
- The facet approach of the instruments themselves. If you use the JSS, its nine facets are your framework for Chapter 2; you do not need a separate theory to justify measuring pay and supervision separately, only a citation to the instrument’s development.
Do not list five theories because they appear in textbooks. Choose the two that your hypotheses actually test, and show the arrow from each theory to a variable. The arrows become the diagram in your conceptual framework, which is built in our guide to the conceptual framework for an MBA dissertation.
Typical designs for a job satisfaction dissertation
Four designs cover the studies you are likely to propose on this construct:
- Descriptive survey of one organisation or sector. Measure facet satisfaction and report means. The weakest design on its own, but acceptable as the first objective of a larger study.
- Comparison across groups. Compare satisfaction by gender, age band, designation, experience band or department. Group variables need operational definitions with their boundaries stated.
- Relationship study. Correlate satisfaction with another construct such as work-life balance, organisational commitment, leadership style, training or intention to leave, then regress if the sample supports it.
- Mediation or moderation. Test whether satisfaction explains the effect of one variable on another. This is the strongest design and the most demanding; it needs a larger sample and a pre-specified model.
Your sample size follows from the design, not from habit. Calculate it from your analysis plan and the tests you intend to run, and state the method you used in Chapter 3.
Hypotheses you can test: an illustrative set
The following are templates, not findings. Replace the bracketed items with your own sector, instrument and sample.
| Objective | Null hypothesis | Likely test |
|---|---|---|
| To assess facet satisfaction among [employee group] in [organisation or sector] | No hypothesis needed for a purely descriptive objective | Means and standard deviations by facet |
| To compare satisfaction between [two groups] | H0: There is no significant difference in overall job satisfaction between [group A] and [group B] | Independent-samples t-test, or Mann-Whitney U if scores are not normal |
| To compare satisfaction across [three or more designations] | H0: There is no significant difference in the [facet] score across designations | One-way ANOVA, or Kruskal-Wallis |
| To examine the relationship between [work-life balance] and satisfaction | H0: There is no significant relationship between [work-life balance] and overall job satisfaction | Pearson or Spearman correlation |
| To determine which facets predict overall satisfaction | H0: None of the facet scores significantly predicts overall satisfaction | Multiple regression, with assumption checks |
Choosing between the parametric and non-parametric versions is a decision you must justify from your data, and the full decision table is in our guide to which statistical test to use. Mapping each objective to a variable, an indicator and an item is the job of an operationalisation table in Chapter 3, so build one before you finalise the questionnaire.
Ten researchable titles to adapt
Bracketed items are yours to fill in with a real sector, city and group.
- Pay, supervision and job satisfaction among [IT, banking or insurance] employees in [city]: a facet-level study
- Work-life balance as a predictor of job satisfaction among [women employees] in [sector]
- Job satisfaction and intention to leave among [mid-level managers] in [industry]
- The effect of [training opportunities] on job satisfaction in [organisation type]
- Does leadership style explain job satisfaction? A study of [sector] supervisors and subordinates
- Job satisfaction across [permanent and contractual] employees in [public or private sector undertaking]
- Satisfaction with fringe benefits and contingent rewards among [frontline staff] in [retail or hospitality]
- Organisational commitment as a mediator between job satisfaction and [performance rating]
- Generational differences in job satisfaction among [employees of a named sector]
- Job satisfaction of [remote or hybrid] employees compared with on-site employees in [sector]
Check each against the existing dissertations in your department before you commit, and ask your guide whether a similar title has been submitted in the last few batches.
Access: the practical problem under every HR dissertation
Satisfaction data comes from employees, which means you need the employer’s permission, a method of reaching respondents that does not identify who answered, and a plan for what you do if the organisation withdraws. If you are writing on a company you intern at, read our guide on what to do when a company will not share HR data before you finalise the topic. A sector-wide study that uses an online questionnaire sent through professional networks is a common fallback, but it changes your sampling claim: say “convenience sample” and do not generalise.
Reliability and what to report
If you adopt a published scale, report the reliability you obtained in your own sample, not only the figure from the original paper. For internal consistency that means Cronbach’s alpha for the total and for each facet you analyse. Facets with few items often produce lower alpha, and you should say so rather than hide it. What counts as acceptable, and what to write when a facet falls short, is covered in our guide to the acceptable Cronbach’s alpha for a thesis.
Five mistakes that send a job satisfaction chapter back
- Using “job satisfaction” and “motivation” as synonyms. Satisfaction is an attitude; motivation is a force toward action. Define both if both appear.
- Averaging all items without reverse-scoring. The JSS, for example, has reverse-scored items. Score according to the instrument’s instructions and say that you did.
- A facet list not tied to a source. “Pay, work environment, growth and recognition” is a list, not a model, unless it is taken from a named instrument or theory.
- Generalising from a convenience sample. Name your sampling method in Chapter 3 and limit your claims to match it.
- Reporting only the total. A total hides the finding. Report facets first.
Writing it up without losing the thread
A job satisfaction dissertation has many moving parts: the definition, the instrument, the group variables, the hypotheses and the tests, and each has to agree with the others. Tesify helps you structure each chapter and assists you as you write, so you can get the skeleton down quickly. More than 9,000 students have used it, it has supported 15,000+ chapters, and every word of your dissertation stays 100% written by you.
Start your dissertation in Tesify
Frequently asked questions
What is job satisfaction in a dissertation?
It is a positive or negative attitude toward one’s job. A common definition is Locke’s (1976): a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences. In your dissertation you pair a conceptual definition like this with an operational one that names your scale and cut-offs.
Which scale should I use to measure job satisfaction?
The Job Satisfaction Survey (Spector, 1985) is a common choice for MBA dissertations because it measures nine facets with 36 items and its author permits use on conditions of attribution and sharing of results. The Job Descriptive Index and the Minnesota Satisfaction Questionnaire are alternatives, but check their permission terms first.
How many facets does the Job Satisfaction Survey have?
Nine: pay, promotion, supervision, fringe benefits, contingent rewards, operating procedures, coworkers, nature of work and communication, with four items for each facet.
Which theories should I use for a job satisfaction dissertation?
Two are enough if each is linked to a variable: Herzberg’s motivator-hygiene theory for intrinsic and extrinsic factors, and Locke’s range-of-affect theory for the gap between wants and outcomes. Add the facet structure of your instrument for the measurement side.
What is a good sample size for a job satisfaction study?
It depends on the design and the tests, not on the topic. A descriptive study needs a different number from a regression with several predictors. Calculate it from your analysis plan and state the method in Chapter 3.
Can I study job satisfaction in one company?
Yes, as a case study, provided you have written permission and you do not generalise beyond that organisation. Make sure respondents cannot be identified from their designation and department together.
What statistical tests are used for job satisfaction?
Group comparisons use t-tests or ANOVA, or their non-parametric equivalents; relationships use correlation and regression; mediation uses a pre-specified model. Choose after checking the distribution of your scores.
How do I write the operational definition of job satisfaction?
Name the instrument, the number of items, the response format, how the total is computed, including reverse-scoring, and any cut-off you use to classify respondents as satisfied or dissatisfied. State where each cut-off comes from.
Is job satisfaction the same as employee engagement?
No. Satisfaction is an attitude about the job; engagement is usually defined as an energetic involvement in work. They correlate, but they are different constructs and need different instruments.
How do I handle reverse-scored items?
Recode them before computing totals, following the scale’s own scoring instructions, and state in Chapter 3 that you did so. Failing to recode is a common cause of low reliability figures.