(Excerpts From) Investigating Performance: Design and Outcomes With Xapi
By Janet Laane Effron and Sean Putman
()
About this ebook
Related to (Excerpts From) Investigating Performance
Related ebooks
Building Big Data Applications Rating: 0 out of 5 stars0 ratingsThriving in a Data World: A Guide for Leaders and Managers Rating: 0 out of 5 stars0 ratingsData Culture A Complete Guide - 2020 Edition Rating: 0 out of 5 stars0 ratingsPractical Enterprise Data Lake Insights: Handle Data-Driven Challenges in an Enterprise Big Data Lake Rating: 0 out of 5 stars0 ratingsData Analysis and Harmonization: A Simple Guide Rating: 0 out of 5 stars0 ratingsData And Analytics Capabilities A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsCenter of Excellence A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsEnterprise Architecture Governance Standard Requirements Rating: 0 out of 5 stars0 ratingsSpark SQL A Complete Guide Rating: 0 out of 5 stars0 ratingsCenter Of Excellence A Complete Guide - 2021 Edition Rating: 0 out of 5 stars0 ratingsDataOps A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsAI Strategy A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsWorkforce Analysis A Complete Guide - 2020 Edition Rating: 0 out of 5 stars0 ratingsSingle customer view Second Edition Rating: 0 out of 5 stars0 ratingsData Platform CDP A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsIntroduction to Reverse Logistics Rating: 0 out of 5 stars0 ratingsData Governance and Data Management: Contextualizing Data Governance Drivers, Technologies, and Tools Rating: 0 out of 5 stars0 ratingsSmart manufacturing A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsMaking Big Data Work for Your Business: A guide to effective Big Data analytics Rating: 0 out of 5 stars0 ratingsData Catalog Third Edition Rating: 0 out of 5 stars0 ratingsTaming The Big Data Tidal Wave: Finding Opportunities in Huge Data Streams with Advanced Analytics Rating: 4 out of 5 stars4/5Analytics and Business Intelligence A Complete Guide Rating: 0 out of 5 stars0 ratingsMachine MLOps A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsBig Data Architecture A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsFostering Innovation: How to Build an Amazing IT Team Rating: 0 out of 5 stars0 ratingsData Visualization Strategy Standard Requirements Rating: 0 out of 5 stars0 ratingsData Management Platform A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsSupply Chain Execution Predictive Analytics Second Edition Rating: 0 out of 5 stars0 ratingsFederal Data Science: Transforming Government and Agricultural Policy Using Artificial Intelligence Rating: 0 out of 5 stars0 ratings
Business For You
The Richest Man in Babylon: The most inspiring book on wealth ever written Rating: 5 out of 5 stars5/5Law of Connection: Lesson 10 from The 21 Irrefutable Laws of Leadership Rating: 4 out of 5 stars4/5Becoming Bulletproof: Protect Yourself, Read People, Influence Situations, and Live Fearlessly Rating: 4 out of 5 stars4/5Crucial Conversations: Tools for Talking When Stakes are High, Third Edition Rating: 4 out of 5 stars4/5Crucial Conversations Tools for Talking When Stakes Are High, Second Edition Rating: 4 out of 5 stars4/5Collaborating with the Enemy: How to Work with People You Don’t Agree with or Like or Trust Rating: 4 out of 5 stars4/5Summary of J.L. Collins's The Simple Path to Wealth Rating: 5 out of 5 stars5/5Set for Life: An All-Out Approach to Early Financial Freedom Rating: 4 out of 5 stars4/5Capitalism and Freedom Rating: 4 out of 5 stars4/5How to Get Ideas Rating: 5 out of 5 stars5/5Leadership and Self-Deception: Getting out of the Box Rating: 4 out of 5 stars4/5Tools Of Titans: The Tactics, Routines, and Habits of Billionaires, Icons, and World-Class Performers Rating: 4 out of 5 stars4/5High Conflict: Why We Get Trapped and How We Get Out Rating: 4 out of 5 stars4/5Just Listen: Discover the Secret to Getting Through to Absolutely Anyone Rating: 4 out of 5 stars4/5Lying Rating: 4 out of 5 stars4/5The Intelligent Investor, Rev. Ed: The Definitive Book on Value Investing Rating: 4 out of 5 stars4/5Nickel and Dimed: On (Not) Getting By in America Rating: 4 out of 5 stars4/5Confessions of an Economic Hit Man, 3rd Edition Rating: 5 out of 5 stars5/5Buy, Rehab, Rent, Refinance, Repeat: The BRRRR Rental Property Investment Strategy Made Simple Rating: 5 out of 5 stars5/5The Five Dysfunctions of a Team: A Leadership Fable, 20th Anniversary Edition Rating: 4 out of 5 stars4/5The Catalyst: How to Change Anyone's Mind Rating: 4 out of 5 stars4/5Suddenly Frugal: How to Live Happier and Healthier for Less Rating: 3 out of 5 stars3/5Your Next Five Moves: Master the Art of Business Strategy Rating: 5 out of 5 stars5/5Carol Dweck's Mindset The New Psychology of Success: Summary and Analysis Rating: 4 out of 5 stars4/5How to Write a Grant: Become a Grant Writing Unicorn Rating: 5 out of 5 stars5/5Company Rules: Or Everything I Know About Business I Learned from the CIA Rating: 4 out of 5 stars4/5Summary of Eve Rodsky's Fair Play Rating: 2 out of 5 stars2/5
Reviews for (Excerpts From) Investigating Performance
0 ratings0 reviews
Book preview
(Excerpts From) Investigating Performance - Janet Laane Effron
data.
1.1: Quantitative and Qualitative Data
When we think of data, we typically think of numbers. Height and weight, miles-per-gallon, ticket resolution time, test scores, time-to-completion. Big Data sounds like the Matrix, like a stream of endless numbers. Numbers are only part of the story, however. In the chapters ahead we’re going to take a close look at both quantitative and qualitative data. Taken together, they provide the two sides of good (meaningful) analysis.
Quantitative data is about numbers
Quantitative data can be counted or measured. It’s the kind of data we see when we look at:
Test scores (raw numbers, pass/fail)
Course related statistics (number of posts, learning objects completed)
Performance measures (sales data, error rate, customer satisfaction)
Benchmarking (gap reduction, goals met)
Numbers are useful and important, but numbers can be limiting when we seek to answer key questions around learning, such as:
Did it work?
What made it work?
Who didn’t it work for? Why not?
The questions that drive an effective learning design are questions that go beyond linear regression to tell the whole story. They try to step beyond correlation and into causality; they try to establish not just the what,
but the why.
Qualitative data is descriptive
Qualitative data concerns that which cannot be measured by a meter stick, a test score, or a time on the clock. It can include:
user feedback
sentiment
content
demographics
Qualitative data puts metaphorical flesh on the bare bones of quantitative data. For example, if we look at text responses to a question, we can measure the word count but that won’t tell us all we need to know about the quality of the responses. Analyzing the content of the responses to that same question will give us qualitative data: a sense of keywords, the degree of understanding, or the application of a concept. Or, in another scenario, we may receive a lot of positive user feedback after a course (qualitative data), but that data doesn’t tell us the whole story. Users may say they liked the course, but we will need some quantitative (and possibly qualitative) performance data to determine if it was an effective course.
In later chapters we’ll explore how to work with both quantitative and qualitative data and look at some basic examples of using them together in analysis. Before we think about using data, however, we need to be sure our data is capable of doing what we need it to do. We need to talk about data quality.
1.2: Data Quality
In an ideal world, all our data would be complete, consistent, readily available, and accurate. When using two (or more!) data sets together, those data sets would have perfectly matched keys,
that is, a common set of values, such as user identification, that allow us to connect data across different data sets. And of course we’d love for our data to have consistent formatting, with no duplicate or incomplete records. In reality, this isn’t the case, and so when we start to do analysis, we first have to spend time finding what’s inconsistent and what’s missing. There’s always some data cleaning
to be done.
Data quality refers to how fit a data set is for its intended purpose, according to characteristics of completeness, validity, accuracy, consistency, availability and timeliness
(ISO 9001). In practical terms, this means we need to look at the data and see if it represents the real world accurately. We need to make sure that different data sets are aligned. In other words, do the data sets represent the same elements consistently?
For example, we may want to know how much time users spend during each visit to our eLearning portal. When we examine our data, we might note that the majority of visits last between 15 and 30 minutes, but one or two visits clock in at, say, 15 hours. Since it’s unlikely that anyone has spent 15 hours continuously doing training, this alerts us that there’s likely a problem in the data somewhere. Whether the problem lies in how we are set up to log user time on the portal, or something else, it’s worth checking into.
When we’re working with multiple data sets, data consistency is especially necessary. The data might be from different systems, or it might be data in the same system over time, and it’s not uncommon to find different date or other numeric formats, different user IDs for the same person, or even different criteria for a field (for example, how we define active user
or time spent on portal
or even pass
).
There’s no getting around the fact that we need to take a hard look at our data before analysis. We can do this through basic data exploration, like reading the raw data, counting types, and matching key-value pairs. We also need to understand thoroughly where the numbers are coming from in terms of actions, criteria, and which interactions in an application interface create which pieces of data. When we do this, the resulting analysis is likely to be more valid, more useful, and less frustrating. We will also gain knowledge about ways to improve future data collection efforts so that we will generate data that will most effectively meet