META MARKETING ANALYTICS PROFESSIONAL CERTIFICATE

Course 2: Introduction to Data Analytics

Week 1: Working with Data

Coursera Study Guide

Click to Enroll in Coursera Meta Marketing Analytics Professional Certificate

CONTENT

This week you’ll get an overview of the Introduction to Data Analytics Course and then you’ll be introduced to setting Goals, Objectives and Key Performance Indicators for marketing campaigns. The 5 steps of a Data Science Project will be explained with the introduction of the OSEMN cycle framework. You’ll finish out the week seeing a real-life application of each step of the OSEMN cycle.

Learning Objectives

  • Define the OSEMN cycle
  • Apply the OSEMN cycle to a given marketing situation
  • Define “clean” and “dirty” data
  • Vocalize high level insights from data sets
  • Write an appropriate analytics question that includes business goals and objectives as well as KPIs

PRACTICE QUIZ: GOALS AND KEY PERFORMANCE INDICATORS

1. Which of the following is true about having access to data? Choose all that apply.

  • It allows you to track key performance indicators.
  • It is difficult to come by.
  • It helps marketers evaluate the success of advertising. (CORRECT)
  • It helps to plan and forecast the results of marketing action. (CORRECT)

Correct: Correct! Having access to data, and properly analyzing it, helps marketers evaluate the success of their campaigns.

Correct: Correct! Having access to data, and properly analyzing it, can help plan and forecast marketing activities and their results.

2. Before obtaining data (the “O” in the OSEMN process), which should be the very first step in the cycle?

  • Come up with a marketing plan.
  • Determine the objective of the analysis. (CORRECT)
  • Gather your sample of 1,000 people.
  • Seek out investors.

Correct: Correct! Without an objective, you won’t know how to sample or meaningfully analyze your data.

3. What does the R in the SMART acronym stand for?

  • Relax
  • Relocate
  • Relevant (CORRECT)
  • Renovate

Correct: Correct! It’s important to make sure the goals are relevant to the greater business strategy.

4. True or false: KPIs are measurable values that can help you track your progress towards your goal.

  • True (CORRECT)
  • False

Correct: Correct! KPIs are usually quantitative, are directional, and have a direct relationship to your goal.

PRACTICE QUIZ: THE OSEMN PROCESS

1. Looking at some data about user profiles for a website, we notice that a relatively large portion of the users (25%) have their age listed as over 100 years old. Is this dirty data? (Select the most appropriate answer.)

  • It IS dirty data because there’s no way that many users are over 100 so it can’t be accurate.
  • It’s NOT dirty data because people can be over 100 years old.
  • It’s PROBABLY dirty data but we need to know more about the data before deciding what to do with it. (CORRECT)

Correct: This is the best choice; there’s probably some kind of error where users are over 100. For example, maybe the age was calculated incorrectly from their birth year. However, we can’t completely rule out the fact that maybe the users tend to be much older than a typical user base. We should investigate but proceed with the expectation that we’ll have to do some data cleaning.

2. Another popular framework for data analytics projects is the Cross Industry Standard Process for Data Mining (CRISP-DM). It has six stages: Business Understanding, Data Understanding, Data Processing, Modeling, Evaluation and Deployment. The CRISP-DM phase of Evaluation is likely closest to which stage of the OSEMN framework?

  • Scrubbing
  • Obtaining
  • Exploring
  • Modeling
  • INterpret (CORRECT)

Correct: Correct! Interpreting data is the culmination of the OSEMN process where you evaluate the analysis you have conducted. This evaluation then leads to business recommendations.

3. An online bookstore wants to analyze what book titles they should advertise to drive more sales. They’ve already pulled their historic sales and web traffic data as well as book title details. What is the MOST likely best next step based on the OSEMN framework?

  • Start creating models that predict the amount of sales if different book titles were promoted.
  • Begin digging into the details of the data and start looking at statistics such as which book genres were most popular.
  • Look for more data about what past customers loved about their book purchases. Possibly even surveying the customers since the business doesn’t have this information already available.
  • Ensure that the data gathered is generally “clean” & well-formatted and make changes as necessary. (CORRECT)

Correct: That’s right! Since we probably already have a good amount of relative data, we want to make sure that we’ve processed it well-enough for the next analysis steps.

GRADED QUIZ: AN ANALYTICS CASE STUDY

1. You’ve been asked to conduct an analysis on customer interests to identify other product offerings these consumers might be interested in. Using the OSEMN framework what would be a sensible approach?

  • Ask questions, find the largest and smallest statistics, go to market
  • Get the data, ensure it is accurate and well-formatted, conduct statistical analysis, generate predictions and insights, translate results to tell a story. (CORRECT)
  • Analyze statistics, build models, focus on production
  • Mail out surveys, read the results, form an opinion

Correct: Great! You basically described the whole OSEMN framework!

2. Which stage of the OSEMN process works to generate predictions and insights from the data?

  • Scrubbing
  • iNterpreting
  • Modeling (CORRECT)
  • Exploring

Correct: Correct! Modeling has two goals: to generate predictions and insights. Models can automate decision making and also help you understand structural patterns.

3. An analyst notices that the data they are working with contains many duplicate records. In which stage of the OSEMN process would they fix these duplicate records?

  • Scrubbing (CORRECT)
  • Interpreting
  • Exploring
  • Obtaining

Correct: Correct! Scrubbing the data cleans raw, often dirty data. This includes removing duplicates.

4. During the Obtain step of the OSEMN framework, you would collect data for the next steps. What do we call data that has been collected by an entity that doesn’t have a clear relationship with its audience?

  • 4th party
  • 3rd party
  • 1st party (CORRECT)
  • 2nd party

Correct: Correct! 1st party data are data that come directly from an entity’s audience.

5. True or False: Creating visualizations of your data is another way to explore your data.

  • F​alse
  • T​rue (CORRECT)

Correct: Correct! Visualizations can help you better understand your data that could be difficult to learn from just the raw numbers and associated statistics.

6. True or False: The iNterpret stage of the OSEMN framework is where you’ll explain your findings to your stakeholders from earlier stages?

  • True (CORRECT)
  • False

Correct: Correct! In the iNterpret stage you take all your findings from earlier stages and package them into an understandable explanation to your stakeholders.

7. What does the M in the SMART acronym stand for?

  • Money
  • Measurable (CORRECT)
  • Modest
  • Meticulous

Correct: Correct! The “M” in SMART is to remind us to create goals that can be measured so we know when they are complete.

8. KPIs are measurable values that can help you track your progress towards your goal.

  • False
  • T​rue (CORRECT)

Correct: Correct! KPIs are measurable values that are meant to help you track your progress towards your goal.

9. Select the best match to complete the following statement:

After modeling, the next stage of the OSEMN process is ________ .

  • Exploring
  • Naming
  • Scrubbing
  • Interpreting (CORRECT)

Correct: Correct! After building a model, or models, it is time to interpret your results. This is where you translate your analysis back into business recommendations and tell a story.

10. Which stage of the OSEMN process has the goal of trying to understand the data as a whole and learn the patterns in the data?

  • Modeling
  • Scrubbing
  • iNterpreting
  • Exploring (CORRECT)

Correct: Correct! Exploring the data is very important and allows you to get a better understanding of your data beyond individual points by observing patterns in and statistics about the data.

11. The iNterpret stage of the OSEMN framework is the last stage of the process. What is the primary goal of a data analysis project?

  • Collect as much data as possible.
  • Create beautiful visualizations.
  • Create and use the most accurate models with your data.
  • Generate insights and understanding. (CORRECT)

Correct: Correct! Generating insights and understanding is the main goal.

12. Mark all that are true about KPIs values.

  • KPIs values are only used if you have a difficult business problem
  • KPIs values are measurable (CORRECT)
  • KPIs values directly relate to your goal (CORRECT)
  • KPIs values are directional (CORRECT)

Correct: Correct! By having measurable values, it makes it easier to understand.

Correct: Correct! KPI values should relate directly to our goal.

Correct: Correct! KPI values are directional (go and down).

13. KPIs are measurable values that are suggested targets but don’t usually align with your goals.

  • Third party data is gathered about a consumer’s browsing habits without them knowing it
  • Third party data is information shared between trusted partners
  • First party data is available to anyone who owns or manages a website (CORRECT)

Correct: Correct! KPIs are measurable values that are meant to help you track your progress towards your goal.

14. What does the T in the SMART acronym stand for?

  • Timely (CORRECT)
  • Tailored
  • Total
  • Temporary

Correct: Correct! The “T” in SMART is to remind us to create goals that can be completed within a reasonable amount of time.

15. During the Explore step of the OSEMN framework, you attempt to get familiar with your data. Which of the following is NOT a way to examine your variable distributions and relationships?

  • Create a statistical model to make predictions about our data (CORRECT)
  • Observes the minimum and maximum values of numerical columns
  • Observed how common the top values are in categorical columns
  • Create visualizations

Correct: Correct! This step is more likely to be done later in the Modeling step instead of the Exploring step.

16. An analyst has been working with their clients to define business goals and identified questions to answer. They are now ready to dive into the OSEMN process. Which stage of the OSEMN process would likely be the next step for this analyst?

  • Scrubbing
  • Exploring
  • Obtaining (CORRECT)
  • Interpreting

Correct: Correct! The OSEMN process defines obtaining the data as gathering raw data sources themselves which comes after defining your business goals.

17. During the iNterpret stage of the OSEMN framework, you translate your analytical findings back to a business context. Which of the following would you likely do in this stage?

  • Recommend actionable insights to your stakeholders based on what you’ve done during your analysis. (CORRECT)
  • Create a statistical model to make predictions about our data.
  • Collect more data to check that your analysis is correct.
  • Create visualizations to show variable relationships.

Correct: Correct! Recommending actionable insights is an important aspect of the iNterpret stage.

18. Which stage of the OSEMN process works to generate predictions and insights from the data?

  • Modeling (CORRECT)
  • iNterpreting
  • Exploring
  • Scrubbing

Correct: Correct! Modeling has two goals: to generate predictions and insights. Models can automate decision making and also help you understand structural patterns.

19. Here are some examples of goals for an online clothing store that sold goods for about $80,000 in revenue over the past half year. Which ones do you think are SMART?

  • Increase the number of visitors for the website
  • Increase the average checkout value to $100 by the end of the year (CORRECT)
  • Sell clothes for a total of $100,000 in revenue during the second half of this year (CORRECT)

Correct: Correct.  This goal is SMART.  It is specific, measurable (average value of $100), achievable, relevant and time-bound (by the end of the year).

Correct: Correct, this is a smart goal – It is specific, measurable ($100,000 revenue), it is achievable, it is relevant and it is time-bound (second half of this year)

20. What does the acronym OSEMN stand for in data analysis?

  • Orchestrate Scaffold Examine Maximize Numerical
  • Obtain Scrub Explore Model iNterpret (CORRECT)
  • Optimize Search Examine Model Name
  • Organize Size Extrapolate Merge Name

Correct: Correct!

21. Imagine a database containing records of sales from a company’s own website. What kind of data is this?

  • 2nd-Party Data
  • None of the above.
  • 1st-Party Data (CORRECT)
  • 3rd Party Data

Correct: Correct! These are data that the company has collected and stored itself

22. Which of the following is an example of dirty data? Select all that apply.

  • Text Data that are not in English
  • Fields that are blank for some records (CORRECT)
  • Inaccurate data (CORRECT)
  • Data that contain duplicate records (CORRECT)

Correct: Correct! While missing values are common in almost all datasets, they are considered to be dirty data, and often need to be assigned a value in order to apply machine learning algorithms.

Correct: Correct! There’s almost nothing to do if the data are simply incorrect. Inaccurate data are certainly dirty data. If you don’t have confidence in your data, then they are of little value.

Correct: Correct! Data that contain duplicate records skew results and can lead to misleading results. Duplicates must be removed in order to ensure a valid analysis.

23. Which of the following is NOT true about models?

  • Machine Learning models make predictions that are 100% accurate. (CORRECT)
  • All models are wrong, but some are useful.
  • Algorithms are used to train models.

Correct: Correct! This is false. We expect our models to be accurate, but they’ll rarely be perfect (if ever!). Perfect prediction scores like 100% are usually a sign that a mistake was made when building the model.

24. What is the primary goal of a data analysis project?

  • To gather the most data possible
  • To generate insights and understanding (CORRECT)
  • To build the most complex models
  • To build the most accurate models

Correct: Correct! Insights and understanding are the primary goals of data analysis. Numbers, models and data are simply the means to build insights and deepen your understanding

Previous – Course
Next – Week 2

Subscribe to our site

Get new content delivered directly to your inbox.

Quiztudy Top Courses

Popular in Coursera

Liking our content? Then, don’t forget to ad us to your BOOKMARKS so you can find us easily!

Subscribe