In a financial performance project the data question comes before the topic. The dependent variable is a ratio such as return on assets or return on equity, calculated from published financial statements, so a title is only workable if the statements for your firms and years can actually be obtained. Decide the measure, the firms, the period and the source first; the research problem follows.
What Does “Financial Performance” Mean in a Project?
Financial performance has no single definition, which is why you must choose one and say so. Most Nigerian projects use one or more of three families of measure: profitability (how much a firm earns relative to its assets, equity or sales), market value (how the market prices the firm relative to its assets) and, less often, efficiency or liquidity. Cite one or two authoritative sources for the family you choose, then write an operational definition that names the formula, the statement lines used and the years. Our guide to the definition of terms shows how to word both parts.
Which Measure Should You Use?
| Measure | Formula | What it shows | Where the inputs sit |
|---|---|---|---|
| Return on assets (ROA) | Net income divided by average total assets | How profitably a firm uses what it controls; useful for comparing firms in the same industry | Income statement and statement of financial position in the annual report |
| Return on equity (ROE) | Net income divided by average shareholders’ equity | Profit relative to the owners’ stake; equals a return on net asset value | The same statements |
| Tobin’s Q | The market value of a firm’s assets divided by their replacement cost; students commonly use a proxy such as market capitalisation plus total liabilities, divided by total assets, and must state the proxy | How the market values the firm relative to its assets | Annual report plus the share price on the exchange |
| Net profit margin | Profit after tax divided by revenue | How much of each naira of sales becomes profit | Income statement |
ROA is the safest first choice for a student: it is simple, widely understood and uses two published statements. ROE is the usual second measure, and it often moves differently from ROA because it is affected by how much the firm borrows. Tobin’s Q was first introduced by Robin Marris as a firm-level variable in his 1964 book and popularised by James Tobin in 1970, but its proxies vary, so say which one you use. Use average figures where the source of your formula says so, and state whether you used opening, closing or average balances.
A fictitious worked example shows the arithmetic. A firm reports a profit after tax of N8.4 billion. Its total assets were N180 billion at the start of the year and N220 billion at the end, so the average is (180 + 220) / 2 = N200 billion and ROA = 8.4 / 200 = 0.042, or 4.2 per cent. Its shareholders’ equity was N36 billion at the start and N44 billion at the end, so the average is N40 billion and ROE = 8.4 / 40 = 0.21, or 21.0 per cent. If its market capitalisation is N150 billion and its total liabilities are N176 billion (so that equity of N44 billion plus liabilities of N176 billion equals total assets of N220 billion), the proxy for Tobin’s Q is (150 + 176) / 220 = 1.48. These are invented numbers; they show the method only.
Which Theories Frame a Financial Performance Project?
Your independent variables decide the theory. If you test how ownership, board structure or managerial behaviour affect performance, the usual choice is agency theory, which comes from Jensen and Meckling (1976), “Theory of the firm: Managerial behavior, agency costs and ownership structure”, in the Journal of Financial Economics, 3(4), 305–360. If you test how a firm’s resources, capabilities or technology affect performance, the resource-based view is a common choice, and Barney (1991), “Firm resources and sustained competitive advantage”, in the Journal of Management, 17(1), 99–120, is the standard source. Pick one theory, explain its link to your variables and draw your variables into a diagram; our article on the conceptual framework shows how.
Where Does the Data Sit?
For a firm-level project the data is the firm’s published financial statements. Four places are worth knowing.
- The Nigerian Exchange (NGX) website. Under its data menu, the Corporate Disclosures section lists company results, delisted companies and a release calendar, and each company links to a profile page. Start there for listed firms, then open each company’s profile to locate its reports.
- The company’s own website. Most listed companies publish annual reports in an investor or about-us section. When two sources give different figures for the same year, use the audited annual report and say so.
- The Central Bank of Nigeria. Its Annual Statistical Bulletin is the place to look for national financial-sector series that you might use as control variables, such as interest rates and exchange rates. Open the edition you cite and note its year.
- The National Bureau of Statistics. For economy-wide series such as GDP and inflation that may enter your model as controls. Our article on where economics students get data explains what each source actually holds.
Because the figures are public documents, a secondary-data project usually raises fewer access and consent questions than a survey, but you must still cite every source, and you should confirm with your department whether any approval applies. Never state that a figure comes from a source you did not open.
What Design Fits a Financial Performance Project?
The usual design is an ex post facto panel study. Panel data follows several firms over several years, so each firm-year is one observation. The sample size is therefore firms multiplied by years: 12 banks observed for 8 years give 12 × 8 = 96 observations, and you can explain that the number is large enough for the regression you plan because it is more than the number of firms alone. Say which firms you included and why (for example, all listed deposit money banks with complete statements for the period), which you excluded, and why, such as missing years or a merger.
A basic model is ROAit = β0 + β1 Leverageit + β2 Sizeit + εit, where i is the firm and t the year. Each coefficient maps to a hypothesis. Choose between pooled, fixed-effects and random-effects estimation and say how you chose; the Hausman test (Hausman, 1978, Econometrica, 46(6)) is widely used to decide between fixed and random effects. Name the software your department supports and run every table from the same data file.
What Variables Go Into the Model?
| Role | Typical choices | How to define it |
|---|---|---|
| Dependent | ROA, ROE, Tobin’s Q | Formula and statement lines, as above |
| Independent | Leverage, liquidity, capital adequacy, board size, ownership structure, audit quality, ICT spend, corporate social responsibility spend | A ratio or count from the annual report, with the formula stated |
| Control | Firm size, firm age, inflation, exchange rate, GDP growth | Firm size is often the natural logarithm of total assets; state it |
Keep the number of independent variables small, because each one needs a hypothesis, a source and a sentence of justification. If your supervisor wants the full set of ratios worked through a results chapter, our article on Chapter Four of an accounting project shows a ratio analysis written out.
What Data Problems Should You Check Before You Start?
- Units. Statements may be in thousands or millions of naira; convert before you compute.
- Restated figures. A later report may restate an earlier year. Use one rule and state it.
- Year ends. Firms may end their financial year in different months; align or note it.
- Missing years. A firm with a gap in the period needs a rule: drop it or fill it from another report.
- Delisted and merged firms. Leaving out firms that disappeared can bias results towards survivors, so say how you handled them.
- Outliers. Check extreme ratios against the statements before you delete anything.
What Titles Can You Use?
These ten are illustrative starting points. Check with your supervisor that each is allowed and that the statements are accessible for your firms and years. The Banking and Finance topic list in our article on final year project topics in banking and finance shows how topics are grouped by the data they rely on.
- Capital structure and financial performance of listed deposit money banks in Nigeria
- Board size and financial performance of listed manufacturing firms in Nigeria
- Liquidity management and profitability of insurance companies in Nigeria
- Firm size and return on assets of listed consumer goods companies in Nigeria
- Audit quality and financial performance of listed financial services firms
- Corporate social responsibility spending and profitability of listed firms
- Credit risk management and financial performance of commercial banks in Nigeria
- Working capital management and profitability of listed manufacturing firms
- Dividend policy and market value of listed firms on the Nigerian Exchange
- Digital banking investment and financial performance of Nigerian banks
What Mistakes Do Panels Flag?
- Choosing firms before checking that statements exist for every year.
- Not saying which version of a ratio you used: average or closing balances, which proxy for Tobin’s Q.
- Treating a correlation as a cause. An ex post facto design supports association; word the conclusions accordingly.
- Too many variables for the number of observations.
- No source for a figure. Name the report, the page and the year.
- Calling a pooled regression the finding without testing fixed or random effects.
Tesify helps you structure and organise your project, and the text is 100% written by you. Start your project with Tesify — 9,000+ students have written 15,000+ chapters with Tesify.
Frequently asked questions
What is the best measure of financial performance for a student project?
Return on assets is the simplest and the most widely used, because it needs two published statements. Return on equity is the usual second measure, and Tobin’s Q suits projects on market value.
Where can I get the financial statements of Nigerian listed companies?
The Nigerian Exchange website lists company results and has a profile page for each listed company, and most companies also publish annual reports on their own sites. Use the audited annual report where figures differ.
How many years of data do I need?
There is no national rule. Many projects use a period long enough to give a reasonable number of observations; agree it with your supervisor and justify it by the availability of statements.
How many observations does a panel give?
The number of firms multiplied by the number of years. For example, 12 firms over 8 years give 96 observations.
Do I need ethics approval to use annual reports?
Public financial statements usually raise fewer consent questions than surveys, but you should cite every source and confirm with your department whether any approval applies.
Should I use average or closing balances in ROA?
Either can be defended, but you must say which you used and follow the formula in the source you cite. Many sources use average total assets.
What is the difference between fixed and random effects?
They are two ways of handling differences between firms in panel data. The Hausman test is widely used to choose between them, and you should report the choice and the result.
Can I study a single company?
Yes, as a case study over several years, but state the limit that the findings describe one firm and cannot be generalised to the sector.
What theory should I use?
Agency theory suits ownership and governance variables, and the resource-based view suits resources and capabilities. Choose the one that explains your independent variables.