IIBA IIBA-CBDA: Certification in Business Data Analytics (IIBA-CBDA)

Last checked November 10, 2026 against the official IIBA exam page.

1. Question 1Identify the Research Questions

   A regional airline's operations director says the team needs to look into why flights are late. Before any data is gathered, which revision of this request is best suited to guide the analysis?

   1. AFind out everything we can about flight delays across the network this year.
   2. BAre our flights more punctual than those of competitors on every route we currently serve?
   3. CCan we reduce delays by hiring additional ground crew at the busiest airports before the summer schedule begins?
   4. DWhich airports and aircraft types have the most late arrivals, and what changed in the last two quarters?
   Show answer and explanationHide answer

   Correct answer: D

   B is a specific, measurable question that names the dimensions to examine and the time frame, so it can guide data gathering. A is too broad to act on, C assumes a solution before any analysis, and D compares against competitors whose data the operations team is unlikely to have.

2. Question 2Interpret and Report Results

   A regression analysis shows that ice cream sales and drowning incidents rise together across months. What is the most appropriate interpretation?

   1. AThe correlation proves that the two measures have no relationship and can be analyzed separately.
   2. BIce cream sales cause the drowning incidents, so the sales of ice cream should be limited in summer.
   3. CDrowning incidents cause ice cream sales, because people buy ice cream after they finish swimming.
   4. DThe two measures move together, but a shared driver such as hot weather may explain both patterns.
   Show answer and explanationHide answer

   Correct answer: D

   Correlation shows that two variables move together but not that one causes the other, and a common driver such as weather often explains both. The other options assert causation in the wrong direction or wrongly claim that correlation proves independence.

3. Question 3Use Results to Influence Business Decision Making

   A product team wants to know whether a new checkout design increases conversions. The analyst proposes randomly showing half of visitors the new design and half the current design. What is this approach, and why is it valuable?

   1. AA descriptive survey, which records what customers say they think about the design in their own words.
   2. BA forecasting exercise, which predicts future conversion rates from the past trend in visitor numbers.
   3. CA data cleansing step, which removes unusual visitor sessions before the conversion rates are computed.
   4. DA controlled experiment such as an A/B test, where randomization helps isolate the effect of the design.
   Show answer and explanationHide answer

   Correct answer: D

   Random assignment to two versions creates comparable groups, so differences in conversion can be attributed to the design. A survey, forecast, or cleansing step does not isolate the design's effect on behaviour.

4. Question 4Analyze Data

   Which type of analytics answers the question of what is likely to happen next quarter?

   1. APredictive analytics, which estimates likely future outcomes from historical patterns
   2. BDiagnostic analytics, which examines the reasons a particular outcome occurred
   3. CDescriptive analytics, which summarizes what has already happened in a given period
   4. DPrescriptive analytics, which recommends actions to take given a set of possible scenarios
   Show answer and explanationHide answer

   Correct answer: A

   Predictive analytics uses historical patterns to estimate what is likely to happen next. Descriptive analytics summarizes the past, diagnostic analytics explains causes, and prescriptive analytics recommends actions.

5. Question 5Source Data

   A manufacturer wants to combine machine sensor logs, maintenance tickets, and supplier spreadsheets to analyze equipment failures. The sensor logs record timestamps in one time zone, while the tickets use local time. What is the best first step?

   1. ADelete every ticket that was created outside the main plant's time zone so that all records line up.
   2. BStandardize the timestamps to one common time zone and record the conversion rule in the data dictionary.
   3. CAnalyze each source separately and avoid joining the sensor logs with the maintenance tickets at all.
   4. DAsk the sensor vendor to regenerate all historical logs in a new format before any analysis is attempted.
   Show answer and explanationHide answer

   Correct answer: B

   Aligning timestamps to one time zone and documenting the rule keeps the joined data accurate and traceable. Deleting records loses valid data, avoiding the join prevents the failure analysis from using both sources, and waiting on the vendor is unnecessary.