Audit Data Quality and Data Analytics AAIA Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   An auditor is analyzing a dataset of customer transactions to identify potential fraud. The auditor runs a data profiling script and finds that the 'transaction_amount' field has a maximum value of $1,000,000 but 95% of the values are below $1,000. Which data quality assessment method is MOST appropriate to further investigate this anomaly?

   • AData cleansing, by immediately removing all transactions above $1,000 to normalize the dataset.
   • BData quality rules, such as checking that all transaction amounts are positive numbers.
   • CPredictive analytics, by building a model to forecast future transaction amounts based on historical data.
   • DStatistical analysis, such as calculating the standard deviation and using a Z-score test to identify outliers.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   An AI auditor needs to query a relational database to extract records that meet specific criteria, such as all transactions above a certain amount. Which data analytics tool is most appropriate for this task?

   • AIDEA
   • BR
   • CPython
   • DSQL
   Reveal Solution Discussion Select an answer first
4. 18expert · hard

   An audit team is preparing a report on the performance of a customer service chatbot. The analysis shows that the chatbot's 'customer satisfaction score' has a wide variance. The team needs to present this finding to stakeholders who are interested in both the overall trend and the distribution of scores. Which visualization approach is MOST effective?

   • AA detailed table listing every individual satisfaction score.
   • BA single line chart showing the average satisfaction score for each month.
   • CA pie chart showing the proportion of high, medium, and low satisfaction scores.
   • DA box plot showing the median, quartiles, and outliers of the satisfaction scores, alongside a line chart showing the trend over time.
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   An audit team is evaluating a model that predicts inventory demand. The audit objective is to assess whether the model's recommendations for the next quarter are reliable. The team has historical sales data, current inventory levels, and supplier lead times. Which combination of analytics techniques is MOST appropriate to achieve this objective?

   • AUse diagnostic analytics to find the root cause of stockouts, then use descriptive analytics to report the findings.
   • BUse prescriptive analytics to determine the best inventory levels, then use predictive analytics to validate the recommendation.
   • CUse predictive analytics to forecast future demand, then use prescriptive analytics to recommend optimal inventory levels based on the forecast.
   • DUse descriptive analytics to summarize past sales, then use diagnostic analytics to explain why sales were high in certain months.
   Reveal Solution Discussion Select an answer first
6. 20expert · hard

   An audit team is testing the effectiveness of a new fraud detection model. The team has access to a dataset of transactions, including a 'fraud_flag' field. The team wants to test whether the model is correctly identifying fraudulent transactions. Which data analytics procedure is MOST appropriate for this specific test?

   • AAnomaly detection, by looking for transactions that are unusual compared to the rest of the dataset.
   • BTrend analysis, by plotting the number of fraudulent transactions over time.
   • CControl testing, by comparing the model's fraud predictions against the actual outcomes in the dataset to calculate the false positive and false negative rates.
   • DDescriptive analytics, by summarizing the total number of transactions and the number flagged as fraudulent.
   Reveal Solution Discussion Select an answer first