Top Banner
Perceptual Data as a Tool by Rhonda Jindra, Kunal Dash, and Justine Yeo
50

Perceptual Data_04182016

Apr 15, 2017

Download

Documents

Kunal Dash
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Perceptual Data_04182016

Perceptual Data as a Tool

by Rhonda Jindra, Kunal Dash, and Justine Yeo

Page 2: Perceptual Data_04182016

Purpose for Today’s Session

• Understand the importance of perceptual data

• Investigate a variety of tools that support the analysis of perceptual text data

Page 3: Perceptual Data_04182016

Perceptual Data

Page 4: Perceptual Data_04182016

Data Types

Page 5: Perceptual Data_04182016

Why Perceptual Data?• One of the four types of data that should

be looked at for school improvement• Measures attitudes of students, school

staff, and parents toward the school environment (education, social, physical, etc.)

• Reflects reality of the stakeholder and guides behavior

• Without it, data is incomplete and may lead to less than optimal decisions

Page 6: Perceptual Data_04182016

The Perceptual Survey• Questionnaires from Education for the

Future• Questionnaires by group:

•Student (Elementary, Middle, High School)•Staff•Parent (K-8th Grade, High School)

Page 7: Perceptual Data_04182016

The Perceptual Survey: Close-Ended Questions

Page 8: Perceptual Data_04182016

The Perceptual Survey:Open-Ended Questions

Page 9: Perceptual Data_04182016

The Perceptual Report

Page 10: Perceptual Data_04182016

The Current State• Perceptual data also used to improve

data literacy among school staff by having them analyze and interpret graphs in the reports

• However, methods for analyzing perceptual text data have not been explored yet

• Text data provides very rich insights for continuous improvement

Page 11: Perceptual Data_04182016

Resource for Analyzing Perceptual Data

• Strengths; Challenges; Implications; So What, Now What?

• Across Surveys: Agreements and Disagreements (Student, Staff, Parent)

Page 12: Perceptual Data_04182016

Visualizing Text Data

Page 13: Perceptual Data_04182016

Using Word Clouds• Various off-the-shelf tools: www.wordle.com and www.worditout.com • Example from www.worditout.com:

• Step 1: Go to www.worditout.com; more specifically http://worditout.com/word-cloud/make-a-new-one • Step 2: Copy text data from Word, Notepad or Excel into the box titled “Original text”• Step 3: Click “Generate”• Step 4: Repeat until satisfied with design of generated world cloud

• Bigger words signify more frequently occurring words

Page 14: Perceptual Data_04182016

14

Using “Word It Out”

Page 15: Perceptual Data_04182016

15

Generated Word Cloud

Page 16: Perceptual Data_04182016

Sentiment Analysis using “Meaning Cloud”

Page 17: Perceptual Data_04182016

Sentiment Analysis• Analysis of a text to determine the underlying tone• Tones can be positive, negative, neutral or undetermined• Possible to analyze and glean insights from text as data just as much as numbers• A relatively affordable sentiment analysis tool or software which we will discuss is an Excel add in - meaning it can be used to analyze text data laid out in a spreadsheet• Software is called “Meaning Cloud”• Subsequent slides: Use one of “Student”, “Staff” or “Parents” text data and understand results

Page 18: Perceptual Data_04182016

“Meaning Cloud” Sentiment Analysis:A Demo

Page 19: Perceptual Data_04182016

19

Step 1

Page 20: Perceptual Data_04182016

20

Step 2

Page 21: Perceptual Data_04182016

21

Step 3

Page 22: Perceptual Data_04182016

22

Step 4

Page 23: Perceptual Data_04182016

23

Step 5

Page 24: Perceptual Data_04182016

24

GLOBAL SENTIMENT ANALYSIS

Results

Page 25: Perceptual Data_04182016

25

EXPLANATION OF OUTPUTDocument Analysis or Global Sentiment Analysis: How to interpret the output meaningfully?

Polarity It could be one of the following colors:

Agreement Shows the agreement between the polarities detected in the text

Subjectivity Shows the subjectivity value obtained for the text

Confidence Shows the confidence value associated to the polarity detected for the text

Irony Shows if the text is considered ironic

Page 26: Perceptual Data_04182016

26

2ND OUTPUT GENERATED

TOPICS SENTIMENT ANALYSIS

Results

Page 27: Perceptual Data_04182016

27

EXPLANATION OF OUTPUTTopic Analysis or Topics Sentiment Analysis: How to interpret the output meaningfully?

Form The name by which the topic extracted is identified

Topic Category Shows the type of topic extracted, that is, if it's an entity or a concept

Rank Contains the order in which the topics have been detected. It's specific for each type of topic, that is, the first entity detected will be ranked 1, and the first concept will be ranked 1 too

Type Shows the type associated to the topic according to our ontology

Polarity Same as in global sentiment analysis

Sense ID Shows the ID of the topic detected

For more information, you can visit the help page on MeaningCloud

Page 28: Perceptual Data_04182016

28

Couple of things to remember:How can you use the free version of this software?

Page 29: Perceptual Data_04182016

29

Couple of things to remember:Where can you find your activation or API key once you

have installed the add in?

Page 30: Perceptual Data_04182016

Additional Tips, Tricks, and Strategies

Page 31: Perceptual Data_04182016

Other Free Text Mining Tools

• Textalyser: http://textalyser.net/ • Termine:

http://www.nactem.ac.uk/software/termine/

• Weka: http://www.cs.waikato.ac.nz/ml/weka/

• R: https://www.r-project.org/

Page 32: Perceptual Data_04182016

Ways to Increase Response Rates

• Multiple contacts: pre-notification, follow-up

• Incentives: class credit• Personalization: personal salutation in

email to staff• Reduce burden: computers during parent

visits/conferences/events

Page 33: Perceptual Data_04182016

BONUS:Working with Google Maps, Tables, and Charts

Page 34: Perceptual Data_04182016

A Google Advantage• To work with Google Maps and Tables,

you need to have a Google account; if you do not have one, then you need to create one

• Once created, please open www.drive.google.com and log in

• Next steps are shown visually• Reference website:

https://www.geolounge.com/make-map-excel-data-using-google-fusion-tables/

Page 35: Perceptual Data_04182016

35

Page 36: Perceptual Data_04182016

36

Page 37: Perceptual Data_04182016

37

Page 38: Perceptual Data_04182016

38

Page 39: Perceptual Data_04182016

39

Page 40: Perceptual Data_04182016

Setting Up Excel Data to Create a Map

Page 41: Perceptual Data_04182016

Geocoding Excel Data

Page 42: Perceptual Data_04182016

Viewing the Map of Excel Data

Page 43: Perceptual Data_04182016

Customizing the Map

Page 44: Perceptual Data_04182016

Creating a Heat Map

Page 45: Perceptual Data_04182016

Customizing Symbols

Page 46: Perceptual Data_04182016

Customizing Symbols

Page 47: Perceptual Data_04182016

Adding a Legend

Page 48: Perceptual Data_04182016

Purpose for Today’s Session

• Understand the importance of perceptual data

• Investigate a variety of tools that support the analysis of perceptual text data

Page 50: Perceptual Data_04182016

Thank you!

Follow us @NDE_GOV!