Difficulty: Medium
A researcher uses last year’s published census table. What type of data is this?
A Primary data
B Experimental data
C Secondary data
D Unrecorded observation
Answer: C. Secondary data. The census table was collected earlier, so it is secondary data for the new study.
A study needs figures for every household in a small village. Which design fits?
A A convenience sample
B A single-cluster sample
C A volunteer survey
D A census
Answer: D. A census. A census covers every household when the population is small and complete coverage is required.
Why stratify a population before sampling?
A To represent important subgroups
B To select only nearby people
C To avoid defining the population
D To remove every survey cost
Answer: A. To represent important subgroups. Stratification can ensure that important subgroups are represented in the sample.
A researcher randomly selects villages, then surveys households there. Which method is used?
A Simple random sampling
B Cluster sampling
C Quota sampling
D Convenience sampling
Answer: B. Cluster sampling. Villages are clusters; randomly selecting villages is cluster sampling.
A list has 1,000 names. After a random start, every tenth name is chosen. Which method is this?
A Stratified sampling
B Convenience sampling
C Systematic sampling
D Census enumeration
Answer: C. Systematic sampling. Selecting units at a fixed interval after a random start is systematic sampling.
A researcher surveys only shoppers leaving one mall. What is the main concern?
A Every resident has equal chance
B The study is a census
C The frame includes all households
D The sample may be unrepresentative
Answer: D. The sample may be unrepresentative. Mall shoppers may differ from the target population, creating selection bias.
A survey asks, βHow satisfied are you with the fast and friendly service?β What is a concern?
A The wording is leading
B The question is open-ended
C The sample is stratified
D The response period is clear
Answer: A. The wording is leading. Words such as βfast and friendlyβ assume positive qualities and can lead respondents.
A household list omits remote settlements. What type of error is likely?
A Response coding error
B Frame undercoverage
C A leading-question effect
D A high response rate
Answer: B. Frame undercoverage. The frame omits part of the target population, causing undercoverage.
A scale records 65 kg for a known 56 kg weight. What error is shown?
A Nonresponse bias
B Sampling-frame error
C Measurement error
D Seasonal variation
Answer: C. Measurement error. The measuring instrument produces an inaccurate value, which is a measurement error.
Respondents forget purchases made months ago. Which error may result?
A Coverage error
B Random-start error
C Classification error
D Recall error
Answer: D. Recall error. Recall error occurs when respondents cannot accurately remember past events or amounts.
An interviewer changes question wording for different people. What is threatened?
A Response comparability
B Population size
C The sampling fraction
D The published data date
Answer: A. Response comparability. Inconsistent wording can change how people interpret questions and reduce comparability.
Online-only sampling excludes households without internet. What is the issue?
A A census error
B Coverage bias
C An import quota
D A price index problem
Answer: B. Coverage bias. An online frame may omit people without internet access, creating coverage bias.
Which method uses chance to select each sample unit?
A Convenience sampling
B Volunteer sampling
C Probability sampling
D Judgmental sampling
Answer: C. Probability sampling. Probability sampling uses random selection with known chances for eligible units.
A researcher deliberately interviews specialist officials. What method is this?
A Simple random sampling
B Systematic sampling
C Cluster sampling
D Purposive sampling
Answer: D. Purposive sampling. Purposive sampling deliberately selects people with relevant knowledge or characteristics.
Why may a bigger biased sample remain misleading?
A Selection bias can persist
B Large samples guarantee balance
C Sample size removes bad wording
D More responses fix undercoverage
Answer: A. Selection bias can persist. Increasing size does not remove a systematic selection bias.
A study records income in rupees. What kind of variable is this?
A Qualitative category
B Quantitative variable
C Sampling frame
D Observation unit
Answer: B. Quantitative variable. Income is numerical and can be analysed as a quantitative variable.
A survey groups farms as small, medium, or large. What type of data are these groups?
A Continuous measurements
B Sampling intervals
C Qualitative categories
D Response rates
Answer: C. Qualitative categories. These are categorical labels rather than continuous numerical measurements.
A survey records each firm’s sales in March. What data structure is this?
A Longitudinal time series
B A randomised trial
C A sampling frame
D Cross-sectional data
Answer: D. Cross-sectional data. Cross-sectional data describe multiple units at one point or period in time.
A study tracks prices each month for five years. What data structure is this?
A Time-series data
B A cross-sectional sample
C A single census
D A cluster frame
Answer: A. Time-series data. Time-series data record observations over successive time periods.
Why compare a secondary source’s definitions?
A Published data need no review
B Measures may not be comparable
C Definitions always stay identical
D It determines the response rate
Answer: B. Measures may not be comparable. Different sources may define populations or indicators differently, limiting direct comparison.
A table lacks its collection date. What should the researcher do?
A Assume it covers this year
B Use it without qualification
C Find its date before comparing
D Replace it with opinions
Answer: C. Find its date before comparing. The date is needed to judge relevance and align the data with other sources.
A survey asks for sensitive personal details. What is important?
A Publish names beside each answer
B Skip consent procedures
C Collect details unrelated to purpose
D Explain privacy and data handling
Answer: D. Explain privacy and data handling. Researchers should explain the purpose and protect sensitive information.
Why use more than one data source?
A To check patterns across sources
B To guarantee identical results
C To avoid documenting methods
D To replace all primary data
Answer: A. To check patterns across sources. Comparing independent sources can help identify inconsistencies and strengthen interpretation.
The unit of analysis is each school. What does the dataset need?
A One record per question
B One record per school
C One record per interviewer
D One record per district only
Answer: B. One record per school. If schools are the units, information should be recorded and analysed at the school level.
Which source is strongest for a study of current village water access?
A An old national price table
B A company export report
C A recent local household survey
D A decades-old city census
Answer: C. A recent local household survey. A recent survey of local households is directly relevant to current village water access.