About this example
This page is an example short research proposal of the kind submitted for admission to a PhD or DBA programme in finance, or as a first-year progression document. It proposes a comparison of the financial performance of listed and unlisted information technology (IT) and IT-enabled services (ITeS) companies in India using annual-report data. At about 1,000 words in its example parts, it shows how to make the background, objectives and method agree with one another, which is what a reviewer checks first.
It is an illustrative teaching example, not a real student's work, and it contains no data. Do not submit any part of it as your own work; your university's academic integrity rules apply. Read each part of the proposal, then the "What to notice" note beneath it.
The example: background and problem
1. Background
India's IT and ITeS industry ranges from a few very large, exchange-listed groups to thousands of privately held companies, many of them subsidiaries or specialist suppliers. Listed companies are subject to the disclosure rules of the Securities and Exchange Board of India and can raise equity on the stock market; unlisted companies file financial statements with the Ministry of Corporate Affairs but rely mainly on retained earnings, bank credit and owner or parent-company funds. Theories of financing predict that this difference matters. If external equity is costly because of information asymmetry, firms without market access will rely more on internal funds and debt (Myers, 1984); and the separation of ownership and control in widely held listed firms creates agency costs that unlisted, closely held firms largely avoid (Jensen & Meckling, 1976). Cross-country evidence also shows that leverage is associated with size, profitability and asset structure (Rajan & Zingales, 1995).
2. Research problem
Analyses of Indian IT firms that rely on stock-exchange data can, by construction, include listed companies only, because their data are the easiest to obtain. If that is the main evidence base, little is known about whether the profitability, asset use, leverage and liquidity of listed IT firms are typical of the sector, or whether listing status is associated with systematically different financial profiles once firm size and sub-sector are taken into account. This matters for investors and lenders who use listed-firm benchmarks, and for policy on access to capital for smaller technology firms.
The example: aim, objectives and questions
3. Aim and objectives
The aim is to compare the financial performance of listed and unlisted IT and ITeS companies in India and to identify firm characteristics associated with any differences. The objectives are:
- To describe profitability (return on assets, ROA, and return on equity, ROE), asset turnover, leverage and liquidity in listed and unlisted IT and ITeS companies over the study period.
- To estimate the difference in ROA and ROE between listed and unlisted companies after accounting for firm size, age and sub-sector.
- To examine whether firm size, leverage, asset turnover and liquidity are associated with profitability, and whether these associations differ by listing status.
The corresponding research questions are: (RQ1) How do the financial profiles of listed and unlisted IT and ITeS companies differ? (RQ2) Is listing status associated with profitability after adjustment for size, age and sub-sector? (RQ3) Do size, leverage, asset use and liquidity relate to profitability in the same way in both groups?
The example: method
4. Research design and data
The study uses a quantitative, longitudinal design based on secondary data (Saunders et al., 2023). Annual financial statements for the ten most recent complete financial years will be taken from a commercial corporate database that covers both listed and unlisted Indian companies, checked against company filings where values are missing or inconsistent. Listing status will be defined as having equity shares listed on the BSE or NSE in a given year.
5. Sample
The population is IT and ITeS companies classified under the relevant national industrial classification codes. To make the groups comparable, the sample will include companies with total assets above a set threshold in the first year, chosen so that enough unlisted firms qualify; a market-value threshold cannot be used because unlisted firms have no market value. Each listed company will be matched with up to two unlisted companies of similar total assets and the same sub-sector (software services, business-process services, products). A target of 40 listed and 80 unlisted companies gives about 1,200 firm-years if records are complete; attrition and missing years will be reported.
6. Variables and analysis
| Construct | Measure | Role |
|---|---|---|
| Profitability | ROA = net profit / total assets; ROE = net profit / equity | Outcome |
| Listing status | 1 if listed on BSE or NSE in the year, else 0 | Main explanatory variable |
| Firm size | Natural log of total assets | Explanatory and matching variable |
| Leverage | Total borrowings / total assets | Explanatory |
| Asset use | Revenue / total assets (asset turnover) | Explanatory |
| Liquidity | Current assets / current liabilities | Explanatory |
| Age and sub-sector | Years since incorporation; sub-sector category | Controls |
RQ1 will be answered with descriptive statistics by group and year. For RQ2 and RQ3, ROA and ROE will be regressed on listing status and the firm characteristics, with year effects and standard errors clustered by firm, because each company contributes several years (Wooldridge, 2010); interaction terms between listing status and each characteristic will address RQ3. Unit-root tests are not planned: with ten annual observations per firm they have little power and are not needed for the questions asked. Because companies choose whether to list, differences will be interpreted as associations, and the matching and controls are intended to reduce, not remove, selection effects.
7. Ethics and data management
The study uses published company financial data only and involves no human participants. Database licence terms will be followed, raw extracts will not be shared beyond the research team, and the cleaning and analysis code will be kept under version control so that the results can be reproduced. Ethical approval requirements will be confirmed with the university research ethics committee.
8. Timeline and expected contribution
| Months | Activity |
|---|---|
| 1 to 6 | Literature review; finalise variables and matching rules |
| 7 to 12 | Data extraction, cleaning and matching; descriptive analysis (RQ1) |
| 13 to 24 | Regression analysis (RQ2 and RQ3); robustness checks |
| 25 to 36 | Writing, supervision review and submission |
The study is expected to provide the first matched comparison of listed and unlisted Indian IT and ITeS firms over a recent decade, to show whether listed-firm benchmarks are representative, and to test whether established determinants of financial performance operate differently with and without access to equity markets.
How to adapt this to your own study
- Start from a gap you can state in two sentences, with the people or decisions it affects (Luse et al., 2012).
- Write objectives that your data can achieve and give each one a research question.
- Build a variables table and check that every concept in your objectives appears in it.
- Make sure your sampling rule can be applied to every group you compare.
- Choose methods that suit the number of periods and the repeated structure of your data.
- Replace news and blog sources with peer-reviewed, official or industry-body sources.
- State what your design cannot show, such as causal effects of a self-selected status.
Common mistakes in this kind of proposal
- Opening with unsourced or inconsistent industry figures.
- Defining a group in a way that contradicts itself (unlisted firms "on the stock exchange").
- Using a sampling criterion, such as market value, that one of the groups cannot meet.
- Listing variables in the objectives and then excluding them in the method.
- Planning time-series tests that need many more periods than the data have.
- Promising comparisons (with other sectors, or with global markets) for which no data are planned.
- Relying on news articles and blogs as the main evidence base.
| What examiners look for | Where this example does it | Check in your own draft |
|---|---|---|
| A clear, specific problem | Section 2 | Can you state the gap in two sentences? |
| Grounding in theory | Section 1 links listing to financing and agency theory | Is your comparison motivated by theory? |
| Aligned objectives and questions | Section 3, one question per objective | Does each objective have a question and a method? |
| A workable sample | Section 5, size-based rule and matching | Can every group meet your criteria? |
| Appropriate analysis | Section 6, panel regression with clustered SEs | Does the method fit your data structure? |
| Honest limits | Selection on listing stated | Have you said what you cannot conclude? |
| Credible sources | Peer-reviewed and methodological references | Are your key claims cited to reliable sources? |
Key takeaways
- Ground the comparison in theory and a specific gap, not in unsourced figures (Sandberg & Alvesson, 2011).
- Give every objective a research question, a variable and a method.
- Use sampling rules that apply to every group you compare.
- Plan panel methods for repeated company data (Wooldridge, 2010).
- Say plainly what a self-selected comparison can and cannot show.
