IIBA-CBDA practice questions: Identify the Research Questions

5 free questions from 52 on this area · answer and explanation for each · no sign-up

These 5 questions come from the Identify the Research Questions section of our IIBA-CBDA bank (52 questions on this area, which carries 20% of the real exam). Every question is original, with the correct answer explained and linked to the source it is drawn from.

Try 6 free IIBA-CBDA questions → All IIBA-CBDA sample questions

Free: 6 questions in study mode, no card. All 52 Identify the Research Questions questions and the full IIBA-CBDA bank: $15 one-time.

1/5 · Identify the Research Questions

An analyst is cataloguing the research questions for a customer-churn initiative. One stakeholder asks, "What specific actions should we take next quarter to retain our highest-value at-risk customers?" Into which type of analytics question does this fall?

ADiagnostic — it asks why something happened.

BDescriptive — it asks what is happening based on historical data.

CPredictive — it asks what is likely to happen in the future.

DPrescriptive — it asks what action to take to achieve a desired outcome.

Answer: D. Research questions are commonly classified by the type of analytics they require: descriptive (what happened), diagnostic (why it happened), predictive (what is likely to happen), and prescriptive (what should be done). A question seeking recommended actions to optimize an outcome is prescriptive. Recognizing the question type guides the analyst toward the appropriate analytical approach. Source: IIBA Business Data Analytics Guide — Identify the Research Questions

2/5 · Identify the Research Questions · hard

An airline wants to study on-time performance and frames the question: "What factors most influence delays?" During refinement, the analyst realizes the team is split: some assume each row of data should represent an individual flight, others a route, and others an entire travel day at an airport. Resolving this disagreement primarily clarifies which aspect of the research question?

AThe unit of analysis — the entity each observation represents — which determines how data is aggregated and how findings are interpreted.

BWhich predictive algorithm will best handle delay data.

CWhether the data should be stored in a relational or columnar database.

DThe sample size needed to reach statistical significance for the chosen test.

Answer: A. A well-framed research question specifies the unit of analysis — the entity (flight, route, customer, transaction, day) that each data observation represents. Ambiguity here leads to incompatible data preparation and misinterpreted results. Settling whether the team studies flights, routes, or airport-days defines the unit of analysis before sourcing and structuring data. Source: IIBA Business Data Analytics Guide — Identify the Research Questions

3/5 · Identify the Research Questions

A retail chain's executive sponsor states the business problem as: 'Our customer loyalty program is underperforming.' Before any data is sourced, the analytics team needs to convert this into something a data analysis effort can act on. Which of the following best represents the team's most appropriate immediate next step?

AImmediately query the loyalty database and build a dashboard so the sponsor can explore the data interactively

BSelect a predictive modeling algorithm appropriate for churn so results are ready when questions arise

CDocument the data quality issues in the loyalty database so they can be remediated

DDecompose the vague problem statement into specific, answerable research questions that target measurable aspects of loyalty program performance

Answer: D. The core purpose of identifying research questions is to convert a broad, often vague business problem into specific and answerable questions. Only once those questions exist can the team determine what data is needed, which techniques apply, and how results will be reported. Jumping to data, tools, or quality work first is a common failure mode. Source: IIBA Business Data Analytics Guide — Identify the Research Questions

4/5 · Identify the Research Questions

A research question requires linking individual web-session behavior to in-store purchases, but the company has no way to connect anonymous web visitors to identified store transactions. Within the activity of identifying research questions, what is the appropriate response to this finding?

AKeep the question unchanged and let the visualization stage resolve the gap

BProceed with the analysis and assume the missing linkage by randomly matching records

CReport that the business problem cannot be analyzed at all and close the initiative

DReassess the question for feasibility and either narrow it to what available data can support or seek a means to obtain the linking data

Answer: D. Part of identifying research questions is confirming each question is feasible given data that is or can be made available. When required data is unavailable, the question should be reassessed, narrowed, or the data sourced, rather than proceeding on assumptions. Source: IIBA Business Data Analytics Guide — Identify the Research Questions

5/5 · Identify the Research Questions

An operations team frames: "Did the new dispatch algorithm improve delivery times?" A senior analyst replies, "Improve compared to what?" and insists this be resolved before the question is accepted. What is the senior analyst requiring the question to specify?

AThe statistical significance threshold - the p-value cutoff that the change in delivery times must clear before improvement can be claimed

BThe privacy controls - the masking and consent checks that the delivery data must pass before it can be used to measure the improvement

CA baseline or benchmark - the reference point, such as pre-change performance or an industry standard, against which improvement is judged

DThe unit of analysis - a structural choice the team must settle about the delivery records before any improvement is measured

Answer: C. Any research question that asserts change—improve, increase, reduce—needs an explicit baseline or benchmark. Without a defined reference point (prior period, control group, or external standard), 'improvement' cannot be measured and the question is not yet answerable. Source: IIBA Business Data Analytics Guide — Identify the Research Questions

Try 6 more free IIBA-CBDA questions → See the whole IIBA-CBDA bank

Other IIBA-CBDA areas

The same kind of free sample for every other section of the IIBA-CBDA bank:

Source Data49 q Analyze Data49 q Interpret and Report Results51 q Secure, Manage and Document Data49 q

Practise IIBA-CBDA with instant feedback

6 free questions in study mode · no card · full IIBA-CBDA bank with area filters and the error log: $15 one-time · no dumps, no ads.

Start IIBA-CBDA free →