Top 5 Data Analyst Interview Questions

Interviews can induce anxiety and cold feet, especially for a data analyst interview where the interviewer's daily job is to analyze! Tough! The best way to combat pre-interview jitters is to prepare yourself. We at Skill Sigma have curated some data analyst questions, with suitable answers to help you prepare for the interview.

1. Can you share details about the largest data set you've worked with? How many entries and variables did the data set comprise? What kind of data was included?

   The largest data set I have worked with was a collaboration with another team therefore a joint project. The data set had more than a million data sets and 500-600 variables. We had to work with the marketing data which we loaded into an analytical tool to perform EDA.

2. What techniques can be used to handle missing data?

   There are varied methods to handle missing data; some of them are:

   • Dropping variables: Can be used if the proportion of missing data is rather huge and the feature is not of great importance. Not recommended usually because it can overload you with too much information
   • Dropping incomplete rows: Best used when the amount of missing data is small & very random
   • Value Imputation: Estimation of the missing field given other information from the sample. Examples- Mean/median/mode imputation, regression models, KNN and multiple imputations. Considering NA (not available) to be a value

3. What is the difference between data profiling and data mining?

   Data profiling is when an analyst is required to monitor and cleanse the data. Whereas, data mining requires the analyst to identify anomalies, patterns and correlations in large data sets to predict the outcome

4. Can you add 1-100 right now?

   While you can count the numbers in the series as 1+2+3+ which is not what the interviewer is looking for. Here is the formula which is called a series sum the number is multiplied by itself + 1 and the result is divided by 2. n(n+1)/2

5. What are precision and recall?

   Precision and recall are metrics that measure classification performance using their own criteria. Formula,

   Precision = TP / (TP + FP)

   Recall = TP / (TP + FN)

   TP = True Positive, FP = False Positives, FN = False Negatives Therefore precision is correctly classified positive cases over predictive cases & recall is the ratio of correctly classified positive cases overall positive cases. Both are together used in the form of F1 Score: F1 = 2 * Precision * Recall / (Precision + Recall)

A career in data science is agile, fast-paced, impactful and dynamic. Now is the time to upskill yourself and get ready towards a successful career path. Skill Sigma offers Certification courses with projects and internships for students. Know more about our Data Science- AI ML Specialization course here.