NigeriaPolls · Guides

The Polling Glossary: 75 Terms Every Nigerian Should Know

A comprehensive glossary of polling and survey research terms, explained in plain English with Nigerian context. From "margin of error" to "stratified sampling" — understand how polls work and what they really mean.

Download this glossary as PDF

Get the complete glossary in a printable format for offline reference.

Download PDF

Section 1

Sampling Terms

Random Sample

A selection of individuals from a population where every member has an equal chance of being chosen. In Nigerian polling, this often means using enumeration area maps from the National Population Commission to ensure no voter group is systematically excluded. True random sampling is the gold standard for minimizing bias.

Stratified Sampling

A sampling method that divides the population into subgroups (strata) — such as by state, gender, or age — and randomly samples from each. In Nigeria, pollsters typically stratify by geopolitical zone to ensure all six regions are proportionally represented, preventing over-representation of Lagos or Abuja respondents.

Quota Sampling

A non-random sampling method where interviewers select respondents to match predefined quotas (e.g., 50% male, 50% female). While faster and cheaper than random sampling, quota samples risk introducing interviewer bias, as seen in some Nigerian street-intercept polls that oversample urban, educated populations.

Probability Sampling

Any sampling method where each member of the population has a known, non-zero probability of being selected. This includes simple random sampling, systematic sampling, and stratified sampling. Leading Nigerian firms like NigeriaPolls and NOI Polls rely on probability methods for their national tracking surveys.

Cluster Sampling

The population is divided into clusters (often geographic), a random sample of clusters is selected, and all individuals within chosen clusters are surveyed. This method reduces travel costs in large, rural countries like Nigeria but increases sampling error compared to simple random sampling.

Systematic Sampling

A probability sampling method where every nth individual is selected from a list after a random starting point. For instance, selecting every 50th name from a voter register. In Nigeria, this works well when reliable population lists are available.

Oversampling

Deliberately sampling more members of a subgroup than their proportion in the population to ensure sufficient data for analysis. Nigerian polls may oversample minority ethnic groups or smaller states to produce reliable state-level estimates, then statistically weight results back to national proportions.

Weighting

A statistical adjustment applied after data collection to correct for demographic imbalances in the sample. If a Nigerian poll sample has 60% male respondents but the population is 50/50, male responses are down-weighted and female responses up-weighted to restore representativeness.

Sampling Frame

The list or database from which a sample is drawn. In Nigeria, common frames include voter registers, enumeration area maps, phone directories, and online panel databases. The quality of the frame directly affects the poll's accuracy — incomplete frames exclude unreachable populations.

Sample Size

The number of completed interviews or responses in a poll. For national Nigerian polls, a sample of 1,200–2,400 respondents is standard, yielding a margin of error of approximately ±3% at 95% confidence. Larger samples reduce error but increase cost. Read more in our sample size guide.

Population

The entire group of people about whom the poll seeks to draw conclusions. For a Nigerian presidential election poll, the population is typically all registered voters or all adults eligible to vote, not just those who actually turned out.

Margin of Error (MOE)

A measure of the potential variability in poll results due to sampling alone. Expressed as a plus-or-minus percentage, it tells you the range within which the true population value likely falls. A poll showing 45% ±3% means the true figure is probably between 42% and 48%.

Confidence Interval

The range of values within which the true population parameter is expected to fall, given a specific confidence level. A 95% confidence interval of 42%–48% means that if the poll were repeated 100 times, 95 of those intervals would contain the true value.

Confidence Level

The probability that the confidence interval contains the true population value. The industry standard is 95%, meaning there is a 5% chance the true value falls outside the reported range. Higher confidence levels (e.g., 99%) produce wider intervals.

Standard Deviation

A measure of how spread out responses are around the average. Low standard deviation means most respondents gave similar answers; high standard deviation indicates wide disagreement. In Nigerian party identification polls, high deviation often signals a politically fragmented electorate.

Section 2

Statistical Terms

Mean (Average)

The sum of all values divided by the number of observations. When a Nigerian poll reports that the "average" respondent rates the economy 3.2 out of 5, they are using the arithmetic mean.

Median

The middle value when all responses are arranged in order. Unlike the mean, the median is not skewed by extreme values. If household income in a Nigerian community poll ranges from ₦50,000 to ₦5,000,000, the median gives a truer picture of the "typical" household.

Mode

The most frequently occurring response in a dataset. If a Nigerian approval poll finds that 45% of respondents rated the president "fair," and no other rating received more votes, "fair" is the modal response.

Variance

The average of the squared differences from the mean. Variance quantifies how dispersed data points are and is the square of the standard deviation. High variance in Nigerian election polls may indicate genuine volatility or measurement problems.

Correlation

A statistical measure of the relationship between two variables. Positive correlation means both move in the same direction; negative correlation means they move oppositely. Nigerian polls often find a positive correlation between education level and support for certain reform policies.

Regression Analysis

A statistical method for estimating relationships between variables. Pollsters use regression to identify which demographic factors (age, ethnicity, income) most strongly predict voting behavior in Nigeria.

Significance Testing

Statistical tests that determine whether observed differences or relationships in data are likely genuine or due to random chance. A "statistically significant" difference in a Nigerian poll means it probably reflects a real gap, not sampling noise.

P-Value

The probability of obtaining results at least as extreme as the observed results, assuming the null hypothesis is true. A p-value below 0.05 typically means the finding is considered statistically significant in academic and polling research.

Turnout Model

A statistical model predicting which survey respondents are actually likely to vote. Turnout models are critical in Nigerian polls because reported voting intention does not always translate to electoral participation, especially in regions with security concerns.

Likely Voter

A respondent classified as probable to vote in an upcoming election, based on past voting history, stated intention, demographic profile, or engagement indicators. Nigerian pollsters refine likely-voter screens using factors like PVC possession and past turnout in the respondent's LGA.

Cross-Tabs (Cross-Tabulation)

Tables showing how responses to one question break down by responses to another — for example, presidential preference by age group or geopolitical zone. Cross-tabs are essential for understanding who supports whom in Nigerian elections.

House Effect

A systematic tendency of a particular polling organization to produce results that consistently lean toward one party or candidate. House effects can arise from methodological choices, sampling frames, or interviewer training differences.

Response Rate

The percentage of contacted individuals who complete the survey. Low response rates increase the risk of non-response bias. Face-to-face surveys in Nigeria typically achieve 70–90% response rates, while phone polls may reach 30–50%.

Refusal Rate

The percentage of contacted individuals who decline to participate. High refusal rates in certain demographics (e.g., wealthier Nigerians screening unknown callers) can skew sample composition.

Non-Response Bias

Error introduced when people who refuse to participate differ systematically from those who agree. If supporters of a controversial candidate are less willing to disclose their preference, polls will understate that candidate's true support.

Section 3

Methodology Terms

Face-to-Face (F2F) Interview

A survey conducted in person, with an interviewer reading questions to the respondent and recording answers. F2F remains the gold standard for Nigerian national polls due to low internet penetration and high mobile call costs in rural areas. It allows for longer questionnaires and complex visual aids.

CATI (Computer-Assisted Telephone Interviewing)

Telephone surveys where interviewers read from a computer screen and responses are entered directly into software. CATI is faster and cheaper than face-to-face but requires reliable phone access. In Nigeria, CATI works well for urban, educated populations but underrepresents rural voters.

Online Panel

A pre-recruited group of respondents who agree to complete surveys over the internet. Online panels are growing in Nigeria but remain unrepresentative nationally due to the digital divide. They are best used for urban consumer and youth-focused research.

IVR (Interactive Voice Response)

Automated phone surveys where respondents answer questions by pressing keypad buttons or speaking. IVR is extremely low-cost and can reach thousands of Nigerians quickly, but it cannot ask open-ended questions and skews toward phone-owning, urban populations.

SMS Survey

A survey distributed and completed via text message. SMS surveys have high reach in Nigeria due to universal mobile phone ownership but are limited to very short questions and suffer from low response quality.

Mixed-Mode Survey

A survey that combines multiple data collection methods — for example, face-to-face in rural areas and CATI in cities. Mixed-mode designs are increasingly common in Nigerian polling to balance cost, coverage, and quality.

Questionnaire Design

The art and science of writing survey questions. Good questionnaire design avoids leading language, keeps questions simple, randomizes answer order to prevent bias, and pre-tests with real respondents. Poor design — like asking "Don't you agree that Candidate X will bring change?" — produces unreliable data.

Screening Questions

Questions asked at the beginning of a survey to determine whether a respondent qualifies for the study. A Nigerian election poll might screen for registered voters, PVC holders, or adults 18+.

Pilot Test

A small-scale trial of a survey conducted before full fieldwork to identify problems with question wording, flow, or translation. Reputable Nigerian polling firms always pilot-test in multiple languages (Hausa, Yoruba, Igbo, Pidgin, English).

Interviewer Bias

Systematic distortion caused by interviewer characteristics or behavior — including tone of voice, appearance, or how they record answers. In Nigeria, matching interviewers to respondents by gender or ethnicity can reduce this bias.

Social Desirability Bias

The tendency of respondents to give answers they believe are socially acceptable rather than their true opinions. In Nigeria, this may lead to overstated support for anti-corruption measures or understated support for ethno-religious preferences.

Acquiescence Bias

The tendency of some respondents to agree with statements regardless of content. This is more common in cultures with high respect for authority. Skilled Nigerian pollsters counter this by including reversed-worded questions.

Fieldwork

The period during which interviews are conducted. In Nigerian polling, fieldwork typically lasts 1–3 weeks and must account for regional security conditions, market days, and religious observances that affect respondent availability.

Data Cleaning

The process of reviewing raw survey data to remove incomplete responses, detect fraudulent interviews, correct errors, and prepare datasets for analysis. Rigorous data cleaning is essential for credible Nigerian polling.

Quality Control

Procedures to ensure survey data accuracy, including supervisor spot-checks, back-checks with respondents, GPS verification of interview locations, and audio recording review. Top Nigerian firms invest heavily in QC.

Section 4

Political Polling Terms

Approval Rating

The percentage of respondents who approve of a political leader's job performance. Nigerian presidential approval ratings from NigeriaPolls and Afrobarometer are closely watched indicators of governing legitimacy and re-election prospects.

Favorability Rating

The percentage of respondents who view a politician positively, regardless of whether they approve of their job performance. A Nigerian governor might have low approval but high favorability if voters like them personally while disagreeing with specific policies.

Horserace Poll

A poll measuring the relative standing of candidates in an election, named after horse racing odds. The "horserace" framing dominates Nigerian election coverage but can obscure important nuances about undecided voters and issue priorities.

Undecided Voters

Respondents who have not yet made a voting decision. In Nigerian polls, undecided voters often range from 10% to 30% weeks before an election. How pollsters allocate or model undecided voters can significantly shift reported margins.

Ballot Test

A survey question that simulates an actual ballot by listing candidate names and asking respondents whom they would vote for. The ballot test is the standard measure of candidate support in Nigerian election polling.

Head-to-Head

A hypothetical matchup between two specific candidates, used even in multi-candidate races to gauge relative strength. Nigerian runoff scenarios are often tested using head-to-head matchups between the top two first-round finishers.

Exit Poll

A survey conducted with voters immediately after they leave polling stations, asking whom they voted for and why. Exit polls are rare in Nigeria due to security concerns and the complexity of simultaneous federal, state, and legislative elections.

Tracking Poll

A poll conducted repeatedly over time using a consistent methodology to detect trends.NigeriaPolls publishes monthly presidential approval tracking, revealing whether support is rising, falling, or stable.

Benchmark Poll

A comprehensive baseline survey conducted early in a campaign to identify voter concerns, candidate strengths and weaknesses, and messaging opportunities. Benchmark polls are typically larger and more detailed than horserace tracking polls.

Push Poll

Not a legitimate poll at all, but a political dirty trick disguised as research. Push polls spread negative information about an opponent while pretending to gather opinions. Learn more in our detailed guide on push polls.

Internal Poll

A poll commissioned by a candidate or party for private strategic use, not public release. Internal Nigerian polls are often more accurate than public polls because campaigns share private data with their pollsters (e.g., voter file information) that improves modeling.

Swing Voter

A voter who does not consistently support one party and may change their vote based on candidates, issues, or events. In Nigeria, swing voters are often found among young urban professionals, first-time voters, and those without strong ethnic or religious party loyalties.

Base Voter

A voter who reliably supports a particular party or candidate regardless of circumstances. Nigerian parties calculate their "base" vote by state and demographic to determine how many swing voters they need to win.

Spoiler Effect

When a third-party or minor candidate draws enough support from a major candidate to change the election outcome. The spoiler effect is particularly relevant in Nigerian presidential elections, where multiple candidates from similar regions can split vote blocs.

Coattail Effect

When a popular candidate at the top of the ticket helps candidates from the same party win down-ballot races. In Nigeria, presidential coattails can influence National Assembly and governorship results in the same election cycle.

Section 5

Nigerian-Specific Terms

BVAS (Bimodal Voter Accreditation System)

An electronic device introduced by INEC for the 2023 general elections that combines fingerprint and facial recognition to verify voters before they cast ballots. BVAS was designed to eliminate multiple voting and impersonation, and its introduction significantly changed how Nigerians experience Election Day.

PVC (Permanent Voter Card)

The smart card issued by INEC to registered voters, containing biometric data and required for voting. PVC collection rates are a critical metric in Nigerian election polling because only PVC holders can vote. Pollsters often weight or filter by PVC possession to improve accuracy.

INEC (Independent National Electoral Commission)

Nigeria's federal election management body responsible for conducting elections, registering voters, registering political parties, and delimiting constituencies. INEC's voter register and accreditation data are essential reference points for polling methodology.

Polling Unit (PU)

The smallest electoral geographic unit in Nigeria where voting takes place, typically located in a public space like a school, market, or community center. There were over 176,000 polling units for the 2023 election. Pollsters use PU-level data for turnout modeling and result verification.

Ward

A cluster of polling units, typically containing 5–15 PUs. Wards are the smallest administrative units in Nigeria and form the basis for local government representation. Ward-level results are useful for granular analysis of voting patterns.

LGA (Local Government Area)

Nigeria's third-tier administrative division, below state and above ward. There are 774 LGAs nationwide. LGA-level analysis is common in Nigerian polling because it balances granularity with statistical reliability.

Geopolitical Zone

Nigeria is divided into six geopolitical zones: North-West, North-East, North-Central, South-West, South-East, and South-South. These zones are not constitutional entities but are widely used in political analysis and polling because they roughly align with ethnic and religious blocs.

Voter Register

The official list of all citizens registered to vote, maintained by INEC. The 2023 register contained approximately 93.4 million voters. Pollsters use the voter register as a sampling frame, though it may not be perfectly up to date.

Accredited Voter

A voter who has been verified by BVAS at a polling unit on Election Day and is eligible to cast a ballot. The number of accredited voters often differs from the number of votes counted, and this gap is scrutinized by election observers and data analysts.

Turnout

The percentage of registered or eligible voters who actually cast ballots. Nigerian turnout has historically been low by global standards — the 2023 presidential election saw approximately 27% turnout by registered voters, down from 35% in 2019. Low turnout complicates polling because likely-voter models become more consequential.

Godfatherism

A Nigerian political phenomenon where powerful individuals ("godfathers") finance and control politicians, expecting loyalty and influence in return. Godfather dynamics can distort polling because publicly expressed preferences may not reflect private political arrangements.

Stomach Infrastructure

A Nigerian political slang term for the practice of distributing food, money, or goods to voters in exchange for support. The term gained prominence during the 2015 election cycle. Polling about vote buying is sensitive because respondents may underreport their exposure or participation.

Agbado Politics

A satirical term derived from a 2022 campaign remark that became emblematic of populist, agriculture-focused economic messaging. The term illustrates how Nigerian campaign discourse quickly enters popular vocabulary and can shape survey responses.

Emilokan

A Yoruba phrase meaning "It's my turn," famously used in the 2023 presidential campaign to frame a rotational candidacy argument. Emilokan became a powerful narrative device that influenced voter perception and was frequently measured in campaign-period polls.

Obidient

The nickname for supporters of Peter Obi's 2023 presidential campaign, derived from his name. The Obidient movement represented a significant shift in Nigerian electoral politics, mobilizing young, urban, and previously politically disengaged voters — a demographic that many polls initially underestimated.

Frequently Asked

Questions About Polling Terms

What is the most important polling term to understand?

Margin of error is the most critical term. It tells you how much a poll result might differ from the true population value. A poll showing a candidate at 45% with a ±3% margin of error means the true support could be anywhere from 42% to 48%.

How is polling different in Nigeria compared to other countries?

Nigerian polling faces unique challenges including linguistic diversity (over 500 languages), security concerns in certain regions, unreliable phone directories, lower internet penetration affecting online panels, and unique electoral systems like the BVAS verification process.

What does "BVAS" mean in Nigerian elections?

BVAS stands for Bimodal Voter Accreditation System. It is an electronic device introduced by INEC for the 2023 elections that combines fingerprint and facial recognition to verify voters before they cast their ballots, reducing electoral fraud.

What is a "push poll" and how is it different from a real poll?

A push poll is not a real poll at all — it is a deceptive political tactic disguised as a survey, designed to spread negative information about a candidate under the guise of research. Unlike legitimate polls, push polls do not collect or analyze data meaningfully.

Why do Nigerian polls sometimes get elections wrong?

Polls can miss due to several factors: low and unpredictable turnout, social desirability bias (voters hiding true preferences), late-breaking events, sampling difficulties in insecure areas, and the challenge of modeling Nigeria's complex ethnic and religious voting patterns.

How large does a Nigerian poll sample need to be?

For national estimates, 1,200–2,400 respondents is standard, yielding a margin of error of approximately ±3%. State-level polls may need 400–600 respondents per state for reliable estimates. Learn more in our sample size accuracy guide.

Keep Learning

Related Guides

Beginner

Stay Informed with NigeriaPolls

Get the latest polls, election forecasts, and analysis delivered to your inbox. No spam — just data-driven insights on Nigerian public opinion.