Primary Research and Market Intelligence

Introduction and context

Real estate decisions are made under conditions of imperfect information. A developer must estimate demand before construction, an investor must assess competing assets, a lender must evaluate marketability, and a public agency must understand housing needs and stakeholder impacts. Secondary data—such as transaction records, census statistics, planning documents and published reports—provide an essential foundation, but they rarely answer every decision question. Primary research supplies direct evidence from households, tenants, buyers, brokers, developers, investors, public officials and other market participants.

Primary research is the systematic collection of new data for a specified decision or research purpose. Market intelligence is the broader managerial process through which information is collected, validated, interpreted and converted into actionable knowledge. The distinction is important: a survey database is not market intelligence until its quality, relevance and implications have been assessed.

In real estate, primary research is especially valuable because markets are local, assets are heterogeneous, information is often privately held, and observed transactions may not reveal unexpressed demand. However, primary research is vulnerable to sampling error, nonresponse, social desirability, interviewer effects, recall problems and measurement error. Professional practice therefore requires a defensible research design rather than simply “asking the market.”

Learning objectives

By the end of this chapter, the reader should be able to:

1. Design an appropriate sampling plan for real estate market research.
2. Select and combine questionnaires, interviews, focus groups, observation and stakeholder consultations.
3. Construct valid and reliable instruments for collecting market evidence.
4. Conduct competitor and comparable-property studies using transparent criteria.
5. Manage fieldwork, quality control, nonresponse and measurement error.
6. Interpret primary evidence critically and convert it into market intelligence.
7. Apply ethical, legal and governance principles to real estate field research.

1. From a management problem to a research design

A sound study begins with a decision problem, not with a preferred method. “Should we develop a residential project?” is too broad to guide fieldwork. It should be translated into research questions such as:

• Which household segments are likely to rent or purchase within the proposed price range?
• What attributes influence choice of location and tenure?
• Which competing properties attract the target segment, and why?
• What unmet needs exist in terms of unit size, services, accessibility or payment terms?
• How sensitive is demand to price, commute time and amenities?

A research design links the management problem to evidence. It should specify the target population, concepts to be measured, data sources, sampling approach, fieldwork procedures, analysis plan, quality controls, budget and timeline.

A useful distinction is between exploratory, descriptive and explanatory research. Exploratory research is appropriate when the problem is poorly understood; it often uses interviews, observation and focus groups. Descriptive research estimates characteristics such as preferences, vacancy perceptions or intended tenure. Explanatory research investigates relationships, for example whether perceived transport accessibility is associated with willingness to pay. The categories can be combined in a sequential design: qualitative exploration may inform a questionnaire, which can then be used for quantitative estimation.

Units of analysis and target populations

The unit of analysis is the entity about which conclusions are drawn. It may be a household, individual consumer, property, tenant, broker, developer, neighbourhood or transaction. The respondent is the person who provides information, but the respondent and unit of analysis are not always identical. For example, one adult may respond on behalf of a household, while a property manager may report information about a building.

The target population is the complete set of units to which the study intends to generalise. The accessible population is the portion that can realistically be contacted. A sampling frame is the operational list or system from which the sample is selected. In real estate, frames may include address registers, tenant lists, property databases, business directories, brokerage records or digitally recruited panels. Every frame has coverage limitations. A broker database may overrepresent active market participants and exclude informal landlords; an online panel may exclude households with limited internet access.

2. Sampling plans for real estate research

Sampling is the process of selecting a subset of a population to produce evidence about that population. A sampling plan should document:

• the target population and unit of analysis;
• the sampling frame and its coverage limitations;
• the sampling method;
• the sample size and allocation across groups;
• eligibility and screening rules;
• recruitment procedures;
• replacement and callback rules; and
• weighting and estimation procedures, if applicable.

Probability and non-probability sampling

In probability sampling, each eligible unit has a known, non-zero probability of selection. This permits design-based estimation and, subject to assumptions and quality controls, supports generalisation to the target population. Simple random sampling, systematic sampling, stratified sampling and cluster sampling are common forms.

In non-probability sampling, selection probabilities are unknown. Convenience, purposive, quota and snowball sampling can be useful when no complete frame exists or when the purpose is exploratory. They may be efficient for reaching specialised stakeholders, such as institutional investors or planning officials, but sampling error cannot be estimated in the conventional probability-sampling sense. Claims should therefore be limited to the participants studied rather than presented as statistically representative of the entire market.

Main probability designs

Simple random sampling gives each frame unit an equal chance of selection. It is conceptually clear but may be expensive when the population is geographically dispersed.

Systematic sampling selects every kth unit after a random start. If the frame contains N units and n are required, the approximate interval is:

Here, N is the number of eligible frame units, n is the desired sample size and k is the selection interval. The method is appropriate when the frame is ordered without a hidden periodic pattern related to the research variable.

Stratified sampling divides the population into internally more homogeneous groups, such as submarkets, tenure categories or income bands, and samples within each stratum. It improves representation of important groups and may increase precision. Proportional allocation is:

where n_h is the sample size for stratum h, n is the total sample size, N_h is the population size of stratum h and N is the total population size. Proportional allocation assumes that representation in the sample should mirror population shares. Disproportionate allocation may be preferable when a small but strategically important segment requires adequate analysis; weights are then usually needed for population estimates.

Cluster sampling selects groups, such as buildings, streets or neighbourhoods, and then surveys units within selected clusters. It can reduce travel costs but often produces less statistical efficiency because units in the same cluster tend to resemble one another.

Sample size and precision

For a simple random sample estimating a population proportion, an often-used preliminary formula is:

Here, n₀ is the initial sample size, z is the standard-normal critical value for the desired confidence level, p is the anticipated population proportion, and e is the maximum tolerated margin of sampling error expressed as a proportion. The formula assumes a probability sample, independent observations, a valid frame and a proportion estimate under the stated confidence approximation. When p is unknown, p = 0.5 produces the largest required sample under this model.

For a finite population of size N, a finite-population correction can be applied:

This correction is relevant when the sample is a substantial fraction of a known, finite population. It does not correct coverage error, nonresponse bias or poor measurement.

A worked illustration: suppose a study uses 95% confidence, z = 1.96, p = 0.50 and e = 0.05. Then:

The preliminary requirement is therefore approximately 385 completed responses. If the target population is N = 2,000:

Approximately 323 completed responses would be required under the stated assumptions. If the expected completion rate is 40%, the number of contacts required is approximately:

Thus, at least 808 eligible contacts would be planned, subject to additional allowance for ineligible cases and unusable records. These calculations address sampling precision only; they do not guarantee representativeness.

For stratified, cluster or weighted designs, the simple random-sample formula may be inadequate. A design effect can be used when justified:

The design effect reflects loss of statistical efficiency from clustering, unequal weights or other design features. Its value should be estimated or justified rather than invented.

3. Questionnaires: constructing valid evidence

A questionnaire translates abstract concepts into observable responses. Common real estate constructs include housing satisfaction, perceived safety, location convenience, willingness to pay, intention to move, tenure preference and service quality.

Operationalisation

Operationalisation involves defining a construct, identifying its dimensions and selecting indicators. For example, “accessibility” may include travel time to employment, public transport availability, pedestrian quality and access to schools. A single question may be adequate for a factual item, but multidimensional constructs often require several carefully designed items.

A good questionnaire follows a logical sequence:

1. introduction, consent and eligibility screening;
2. easy factual and behavioural questions;
3. substantive evaluations and preferences;
4. sensitive financial or demographic questions; and
5. closing comments and contact or incentive procedures.

Questions should be specific, neutral and understandable. Avoid double-barrelled wording such as “How satisfied are you with the rent and maintenance?” because respondents may hold different views about each. Avoid leading wording, technical jargon, ambiguous time periods and unsupported assumptions. A recall question should identify a reference period, such as “during the past 12 months.”

Response scales must match the construct. Agreement scales measure agreement, not necessarily importance. Importance and satisfaction should not be combined unless the analytical purpose is clear. Balanced response options should include both positive and negative positions where appropriate, as well as “don’t know” or “not applicable” when these are substantively meaningful.

Reliability and validity

Reliability concerns consistency of measurement. A scale can be internally consistent yet invalid if it consistently measures the wrong construct. Validity includes content validity, construct validity, criterion-related validity and, in applied settings, practical validity.

Pilot testing should examine comprehension, response burden, routing, missing data, interviewer interpretation and the distribution of responses. Cognitive interviewing—asking participants to explain how they interpreted questions—can identify hidden ambiguity. Translation and back-translation are important in multilingual markets, but cultural adaptation may require more than literal translation.

A commonly used measure of internal consistency is Cronbach’s alpha:

Here, k is the number of items, σᵢ² is the variance of item i and σ_T² is the variance of the total score formed from the items. Alpha assumes a suitable scale structure and is influenced by the number of items; a high value does not establish unidimensionality or validity. It should therefore be interpreted alongside theory, item analysis and, where appropriate, factor analysis.

4. Interviews, focus groups and stakeholder consultation

Semi-structured interviews

Semi-structured interviews use a consistent topic guide while allowing probing. They are appropriate for understanding motivations, institutional constraints, market narratives and processes that are difficult to observe directly. A strong guide contains opening questions, core topics, probes, transition questions and a closing opportunity for correction or addition.

Interviewers should distinguish facts, interpretations and forecasts. For example, “What was the vacancy rate last quarter?” is different from “Why do you think vacancy increased?” and “What do you expect next year?” Respondents should not be pressured to disclose confidential information. Interviews should be recorded only with consent, transcribed accurately and anonymised where required.

Purposive sampling is often appropriate for interviews. Maximum-variation sampling can include different stakeholder positions, while criterion sampling selects participants who meet a defined condition, such as having managed a comparable asset for at least three years. Saturation is a practical stopping concept: additional interviews produce little new insight relevant to the research question. Saturation is not a mechanical statistical rule and should be documented transparently.

Focus groups

A focus group is a moderated discussion among participants who share relevant characteristics or experiences. It is useful for exploring language, reactions to concepts, trade-offs and social meanings. It is not a substitute for a representative survey. Group dynamics can generate insight but can also suppress dissent, amplify dominant voices or produce conformity.

Groups should be sufficiently homogeneous on characteristics that affect interaction, while avoiding unnecessary sameness. Tenants may be reluctant to criticise a landlord in a mixed group; first-time buyers may speak differently in a group with property professionals. A skilled moderator uses neutral probes, invites quieter participants and prevents one participant from controlling the discussion. Analysis should distinguish widely shared views from isolated but analytically important perspectives.

Stakeholder consultation

Stakeholder consultation is an organised process of obtaining views from parties affected by or capable of influencing a project. Relevant stakeholders may include residents, landowners, tenants, local government, utilities, transport agencies, community organisations, lenders, developers and environmental groups.

Consultation should not be reduced to public relations. A stakeholder map can classify parties by influence, interest, potential impact and knowledge. The process should disclose the purpose of consultation, what decisions are open to influence, how information will be used and what feedback will be provided. Token consultation, in which views are collected after decisions are effectively fixed, undermines trust.

5. Observation and behavioural evidence

Observation records what people do or what conditions exist, rather than relying solely on what respondents say. In real estate, structured observation can assess pedestrian flows, parking utilisation, queue lengths, frontage activation, building maintenance, accessibility, signage, noise and public-space use.

An observation protocol should define the site, observation periods, units to be counted, categories, duration, weather conditions, observer position and recording method. Repeated observations at different times and days are essential because activity patterns vary. Observers should be trained with examples and, where feasible, two observers should independently code part of the material.

Observation does not automatically reveal motives. A quiet public space may reflect poor design, bad weather, safety concerns or the observation time. Privacy is also critical: researchers should avoid unnecessary identification of individuals, comply with applicable law and disclose recording where required.

6. Competitor and comparable-property studies

A competitor study examines assets that compete for a customer, tenant, investor or development opportunity. A comparable-property study identifies assets sufficiently similar to support benchmarking or valuation-related analysis. “Comparable” does not mean identical; it means similar in the attributes relevant to the decision.

Potential criteria include:

• geographic submarket and accessibility;
• asset type, tenure and use;
• completion date and physical condition;
• unit or floor-plate size;
• amenities, management and service quality;
• occupancy, vacancy and absorption;
• asking and achieved rents or prices;
• incentives, payment terms and transaction timing; and
• target customer segment.

A competitor matrix should distinguish verified facts from broker claims and researcher judgments. Data should be collected consistently, ideally using the same observation protocol and date reference. Asking rent is not achieved rent; advertised availability is not necessarily physical vacancy; and a headline sale price may exclude incentives, taxes, fit-out or other adjustments.

A practical comparable-selection process is:

1. define the decision and the relevant market segment;
2. assemble a broad candidate list;
3. screen candidates against explicit inclusion criteria;
4. verify characteristics through documents, site visits and interviews;
5. record differences and adjustment implications;
6. analyse patterns rather than relying on a single comparable; and
7. document exclusions and uncertainty.

For a simple descriptive benchmark, the mean of observed values is:

where x̄ is the arithmetic mean, xᵢ is observation i and n is the number of valid observations. The mean can be distorted by unusually high or low values. The median is often more robust for skewed rents or prices. Any comparison should state whether values are nominal or inflation-adjusted, gross or net, and whether they are per unit, per square metre or another consistent basis.

Comparable analysis is not mechanical valuation. Differences in location, quality, timing, incentives and legal status require professional judgment. Small samples and selective disclosure create substantial uncertainty.

7. Fieldwork management and data quality

Fieldwork converts a research design into observations. A fieldwork plan should specify recruitment scripts, interviewer training, contact attempts, safety procedures, consent, supervision, escalation rules, data-security controls and daily reporting.

Training should cover the purpose of the study, definitions, neutral probing, handling refusals, eligibility screening, skip patterns, distress or sensitive disclosures and technology failures. Interviewers should not improvise changes that alter question meaning. Any necessary amendment should be documented and approved.

Quality assurance may include:

• pilot interviews and pilot site visits;
• real-time checks for impossible values and missing fields;
• back-checks with a subset of participants;
• review of interview duration and location metadata where ethically justified;
• duplicate and fraud detection in online surveys;
• inter-rater reliability checks for observation coding; and
• an auditable version history for questionnaires and datasets.

Data cleaning must not silently remove inconvenient responses. Rules for treating duplicates, outliers, partial interviews and inconsistent answers should be defined before substantive analysis where possible. Corrections should preserve the original record and create a documented derived variable.

8. Nonresponse, coverage and measurement error

Total survey error includes several sources of error. Sampling error arises because a sample rather than the full population is observed. Coverage error occurs when the frame excludes or overrepresents segments. Nonresponse error occurs when participants differ systematically from nonparticipants. Measurement error arises from the question, respondent, interviewer, mode or recording process. Processing error is introduced during coding, entry or transformation.

Nonresponse should be monitored by stage: contact failure, refusal, ineligibility, break-off and unusable completion. A response rate alone does not reveal bias. Researchers should compare respondents with available population benchmarks, analyse response patterns by subgroup and conduct sensitivity analyses where feasible.

Practical remedies include multiple contact attempts at varied times, clear invitations, reasonable incentives, short and accessible instruments, multilingual options, privacy assurances and alternative modes. Weighting can adjust for known differences, but it cannot fully repair unmeasured nonresponse bias. Weighting variables and procedures must be reported.

Measurement error can be reduced through precise wording, appropriate recall periods, neutral administration, visual aids used consistently, interviewer training and pilot testing. Sensitive questions may require self-completion rather than interviewer administration. For willingness-to-pay questions, stated intentions should be interpreted cautiously because hypothetical responses may not predict actual behaviour.

9. Analysis and conversion into market intelligence

Analysis should follow the research question and the data-generating process. Quantitative analysis may include frequencies, cross-tabulations, confidence intervals, subgroup comparisons and regression models. Qualitative analysis may use coding, thematic analysis, constant comparison and negative-case analysis. Mixed-method research should integrate findings rather than placing qualitative and quantitative results in separate, unrelated sections.

Triangulation compares different sources, methods or stakeholder perspectives. Agreement increases confidence, but disagreement is not necessarily failure. A difference between tenant reports and property-manager reports may reveal information asymmetry or different definitions of “maintenance quality.” Contradictions should be investigated and reported.

Market intelligence should communicate:

• the finding;
• the evidence and its quality;
• the degree of uncertainty;
• the likely explanation;
• the decision implication; and
• the recommended action or further evidence required.

A useful reporting discipline is to separate observation from inference and inference from recommendation. “Three of five competing buildings reported occupancy above 90%” is an observation subject to verification. “The submarket is healthy” is an inference requiring broader evidence. “Proceed with a premium project” is a recommendation requiring financial, planning and risk analysis beyond the observation itself.

Illustrative case: Riverside Commons mixed-use development

Background

Riverside Commons is an illustrative case involving a proposed mixed-use project near a metropolitan transit interchange. The sponsor is considering 240 apartments, ground-floor retail and shared workspaces. Secondary data indicate population growth, but competing projects have reported inconsistent leasing performance.

Problem

The sponsor needs to determine the likely target segments, appropriate unit mix, competitive positioning and risks before committing to detailed design. The team is concerned that online interest surveys may overstate actual demand.

Approach

The researchers developed a sequential mixed-method design. First, they conducted 18 semi-structured interviews with renters, brokers, employers, transport officials, nearby residents and property managers. Purposive sampling ensured variation by tenure, income segment and location. Interviews indicated that transport access was attractive, but respondents repeatedly mentioned noise, storage and uncertainty about service charges.

Second, the team carried out structured observations at six competing properties and the proposed site. Each site was visited on weekday morning, weekday evening and weekend periods. The protocol recorded pedestrian activity, parking use, visible maintenance, amenity access and retail occupancy. Two observers independently coded a subset of visits to identify disagreements.

Third, a questionnaire was piloted with 30 eligible renters. Cognitive interviews showed that “affordable” meant different things to participants, so the final instrument asked about specific monthly payment ranges and included service charges. The survey used stratified recruitment across nearby neighbourhoods and current renters in comparable buildings. The researchers clearly labelled results as estimates for the defined accessible population rather than the entire metropolitan market.

Finally, the team built a comparable-property matrix. It separated asking rents from verified achieved rents and recorded incentives, unit size, furnishing, parking and lease terms. The analysis used medians within comparable segments and documented differences rather than applying unexamined averages.

Findings and solution

The evidence suggested three viable segments: transit-oriented young professionals, smaller households seeking predictable monthly costs and selected employers seeking furnished short-term units. Demand was weaker for large premium units than the sponsor’s initial concept assumed. Residents valued transport access but considered acoustic treatment, storage and transparent service charges essential. Competitor occupancy was strong in well-managed buildings but weak where promised amenities were poorly maintained.

The recommended response was to revise the unit mix toward smaller and mid-sized apartments, provide acoustic mitigation and storage, publish a service-charge schedule, phase the workspace component and test retail demand with pre-leasing commitments. The sponsor also commissioned a follow-up pricing exercise using a controlled set of stated-choice scenarios, while recognising that stated preferences were not equivalent to signed leases.

Outcome and learning points

The research did not produce certainty; it reduced avoidable uncertainty and changed the design before substantial capital was committed. The case demonstrates five principles: begin with a decision problem, triangulate methods, define comparable evidence carefully, distinguish stated intention from behaviour, and convert findings into design and risk actions rather than attractive but unsupported conclusions.

10. Ethics, governance and technology

Real estate research often involves financial information, household circumstances, location data and potentially vulnerable participants. Ethical practice requires informed consent, voluntary participation, minimisation of harm, confidentiality, secure storage, limited access and a clear retention period. Researchers must comply with applicable privacy, data-protection, consumer-protection and research-ethics requirements.

Participants should not be misled about the sponsor, purpose or commercial use of information. Incentives must not be so large that they undermine voluntary participation. Conflicts of interest should be disclosed, especially when a developer funds research intended to support an investment decision.

Digital tools can improve fieldwork through mobile data capture, geocoded observation, computer-assisted interviewing, dashboard monitoring and automated transcription. AI can assist with transcription, preliminary coding, translation or anomaly detection, but outputs require human verification. Sensitive data should not be uploaded to systems without appropriate contractual, security and governance controls. Synthetic or automated text must never be presented as a participant’s actual statement.

Contemporary practice increasingly combines administrative records, mobile or sensor data, platform listings and primary research. These sources can improve timeliness but introduce new coverage, consent and representativeness problems. More data does not automatically mean better intelligence.

Best-practice checklist

Before approving a primary real estate market study, confirm that:

1. the decision problem and target population are explicit;
2. the sampling frame and coverage limitations are documented;
3. the method matches the research purpose;
4. the questionnaire or topic guide has been piloted;
5. sample allocation reflects strategic subgroups;
6. interviewers and observers are trained;
7. nonresponse and fieldwork deviations are monitored;
8. comparable properties are selected using transparent criteria;
9. asking, achieved, gross and net measures are not confused;
10. data security, consent and conflicts of interest are addressed; and
11. conclusions distinguish evidence, uncertainty and recommendation.

Chapter Summary

Primary research provides direct evidence for real estate decisions, but its value depends on design quality. A credible study defines the target population, unit of analysis and sampling frame before selecting probability or non-probability methods. Sample-size formulas can support planning under stated assumptions, but they do not correct coverage, nonresponse or measurement problems.

Questionnaires require careful operationalisation, neutral wording, appropriate scales, pilot testing and evidence of reliability and validity. Interviews and focus groups reveal motivations, meanings and institutional processes, while observation captures behaviour and physical conditions. Stakeholder consultation should be transparent and capable of influencing decisions.

Competitor and comparable-property studies require explicit inclusion criteria, consistent measures, verification and careful treatment of differences. Fieldwork quality depends on training, supervision, documentation and data-security controls. Nonresponse, coverage error and measurement error must be monitored rather than hidden. Finally, market intelligence emerges when evidence from multiple methods is triangulated, uncertainty is communicated and findings are connected to practical design, investment and management decisions.

Key Terms

Target population: The complete set of units to which the study intends to generalise.

Sampling frame: The operational list or system from which a sample is selected.

Unit of analysis: The entity about which conclusions are drawn.

Probability sampling: Sampling in which each eligible unit has a known, non-zero selection probability.

Non-probability sampling: Sampling in which selection probabilities are unknown.

Stratified sampling: Sampling that divides a population into groups and samples within each group.

Cluster sampling: Sampling that selects groups of units, such as buildings or neighbourhoods.

Coverage error: Error caused by mismatch between the sampling frame and the target population.

Nonresponse error: Bias arising when respondents differ systematically from nonrespondents.

Measurement error: Error caused by the instrument, respondent, interviewer, mode or recording process.

Questionnaire: A structured instrument for collecting standardised responses.

Semi-structured interview: An interview guided by common topics but allowing probing and flexibility.

Focus group: A moderated discussion used to explore perceptions, language and group reactions.

Stakeholder consultation: A structured process for obtaining views from affected or influential parties.

Comparable property: An asset sufficiently similar to support a specified benchmarking or analytical purpose.

Triangulation: Comparison of evidence across methods, sources or perspectives.

Reliability: The consistency of a measurement procedure.

Validity: The extent to which a measure supports the intended interpretation.

Market intelligence: Validated and interpreted information used to support market decisions.

Design effect: The relative loss of sampling efficiency caused by features such as clustering or unequal weights.

Review Questions

5 Short Answer Questions

1. Distinguish a target population from a sampling frame.
2. What is the main difference between probability and non-probability sampling?
3. Why is pilot testing important in questionnaire design?
4. Identify two advantages and two limitations of observation in real estate research.
5. Why should asking rent and achieved rent be reported separately?

5 Long Answer Questions

1. Explain how a real estate management problem is translated into a research design. Illustrate your answer with a development feasibility example.
2. Compare stratified, cluster and purposive sampling, including the circumstances in which each may be appropriate.
3. Discuss the principles of constructing a valid and reliable real estate questionnaire.
4. Explain how interviews, focus groups and stakeholder consultations differ in purpose, sampling and analysis.
5. Describe a complete fieldwork quality-control system for a multi-site competitor study.

3 Analytical / Application Questions

1. A developer wants to estimate renter preferences across three neighbourhoods, but the available database overrepresents high-income households. Design a sampling and weighting strategy and identify its limitations.
2. A survey reports strong interest in a proposed luxury apartment project, while comparable properties show slow leasing. Develop a triangulation plan to investigate the contradiction.
3. Construct a comparable-property framework for a mixed-use project. Specify inclusion criteria, variables to collect, verification procedures and how uncertainty would be communicated to an investment committee.

Multiple Choice Questions with Answers

Question 1

Which element defines the complete set of units to which a study intends to generalise?

A. Sampling frame

B. Target population

C. Respondent quota

D. Observation protocol

Correct Answer: B. Target population

Question 2

A researcher divides households into income bands and samples within each band. Which sampling design is being used?

A. Cluster sampling

B. Snowball sampling

C. Stratified sampling

D. Convenience sampling

Correct Answer: C. Stratified sampling

Question 3

Using z = 1.96, p = 0.50 and e = 0.05, the preliminary simple-random-sample size is approximately:

A. 96

B. 196

C. 323

D. 385

Correct Answer: D. 385

Question 4

Which statement about Cronbach’s alpha is correct?

A. It proves that a scale measures the intended construct

B. It measures internal consistency under stated scale assumptions

C. It is used only to calculate sampling intervals

D. It eliminates nonresponse bias

Correct Answer: B. It measures internal consistency under stated scale assumptions

Question 5

What is a principal advantage of semi-structured interviews?

A. They guarantee population representation

B. They eliminate interviewer effects

C. They combine consistency with opportunities for probing

D. They require no sampling rationale

Correct Answer: C. They combine consistency with opportunities for probing

Question 6

Which statement best distinguishes a comparable property from an identical property?

A. A comparable property is similar on attributes relevant to the decision

B. A comparable property must be in the same building

C. A comparable property must have the same asking price

D. A comparable property cannot differ in quality

Correct Answer: A. A comparable property is similar on attributes relevant to the decision

Question 7

Which problem occurs when a sampling frame excludes informal landlords?

A. Processing error

B. Coverage error

C. Recall error

D. Interviewer error

Correct Answer: B. Coverage error

Question 8

Why should a researcher distinguish stated intention from observed behaviour?

A. Intentions always have lower measurement cost

B. Stated intentions may not predict actual market actions

C. Observed behaviour cannot be recorded ethically

D. Intentions are always statistically representative

Correct Answer: B. Stated intentions may not predict actual market actions

Question 9

Which practice most directly improves the comparability of site observations?

A. Allowing each observer to create categories during fieldwork

B. Using a common observation protocol and repeated time periods

C. Observing only the most attractive property

D. Removing contradictory observations

Correct Answer: B. Using a common observation protocol and repeated time periods

Question 10

What is the primary purpose of triangulation?

A. To replace all qualitative research with surveys

B. To compare evidence across methods, sources or perspectives

C. To increase the asking rent of a comparable property

D. To guarantee that all stakeholders agree

Correct Answer: B. To compare evidence across methods, sources or perspectives