sadc course in statistics exploratory data analysis for single variables module b2 session 12

19
SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

Upload: megan-cahill

Post on 28-Mar-2015

225 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

SADC Course in Statistics

Exploratory Data Analysis for single variables

Module B2 Session 12

Page 2: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

2To put your footer here go to View > Header and Footer

Learning Objectives

students should be able to

• Explain the importance of exploring data• at the start of the analysis

• Use two new tools for exploration • Dot plots and stem & leaf plots• Construct simple and jittered dot plot• Draw a stem and leaf plot

• Use training resources more effectively• CAST as a training resource• Excel as a training and analysis tool

• With charts and graphs• Explain the difference between exploratory and

presentation graphs

Page 3: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

3To put your footer here go to View > Header and Footer

Stages in processing the data

• Entry and checking the data

• Organising the data for analysis

• Exploring the data

• Analysis

• Reporting

• The middle 3 stages are iterative and can be repeated

• Some exploration can be before the organising• Continue to explore through the analysis

Page 4: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

4To put your footer here go to View > Header and Footer

In this session

• Two new tools are introduced• Dot plots• Stem and leaf plots

• They are to process numeric data

• So far we have concentrated on categorical data

• Now we start to redress the balance

• In the next session• We apply these tools

Page 5: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

5To put your footer here go to View > Header and Footer

Jittered dot plots in CAST and Excel

CAST

EXCEL

Rainfall data: 608, 746, 767, ….. 1395, 1425, 1482

Page 6: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

6To put your footer here go to View > Header and Footer

Stem and leaf plots – survey yields

1 92 4 4 5 6 7 7 93 0 1 3 6 7 7 8 4 0 0 1 2 2 2 4 5 6 8 9 5 0 3 6 7 8 96 1 2

11 92 4 42 5 6 7 7 93 0 1 33 6 7 7 84 0 0 1 2 2 2 4 4 5 6 8 95 0 35 6 7 8 96 1 26

Single stem Split stem

Stem - tens digit

Leaves - units digit

Dec point - truncated

Yields: 19.1, 24.3, 24.7,….. 59.3, 61.4, 62.1

Page 7: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

7To put your footer here go to View > Header and Footer

Exploratory and presentation graphs

• Dot plots and stem and leaf plots • show all the data• to help with exploration• to look for oddities• and to prepare for the analysis

• They are for data exploration• the graphs have to be effective, not “pretty”

• Bar charts and pie charts• show summaries• to present results• to others• in reports and presentations

• They are for presentation

Page 8: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

8To put your footer here go to View > Header and Footer

Practical

• Activity 2 – uses CAST for dot plots

• Activity 3 – uses Excel to produce dot plots

• Activity 4 – uses both for stem and leaf plots

• It also prepares for the future• Efficient use of CAST• Effective use of Excel

Page 9: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

9To put your footer here go to View > Header and Footer

In the practical – what was your response?

1. I was able to see the 2 groups and clicked the answer button just to check

2. I did look for “high and low densities” as it asked, for quite a time, but was not sure what to look for. So I clicked the answer button. Now I will be clearer in the future.

3. I wasn’t sure what to look for, so I pressed the answer button.

4. I didn’t read the instructions carefully enough, but I did click.

5. I went through this page quickly, and didn’t get to this point on the exercise

6. I didn’t do this page

Page 10: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

10To put your footer here go to View > Header and Footer

Learning to use resources fully

• CAST is a new type of resource

• You may have to take some time• to learn how to use it fully• so you gain the maximum from the resource

• When you do• It can help during this training• and also afterwards

• Because it supports self-study

• It is part of an effort to change learning• towards a voyage of discovery

Page 11: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

11To put your footer here go to View > Header and Footer

Using CAST fully?

Follow the instructions to take advantage of the dynamic

elements

Also think why the action is useful

You also saw this earlier

In the tutorial introduction

Page 12: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

12To put your footer here go to View > Header and Footer

Did you puzzle, or just click?

Did you follow these

instructions to scan down the list and look for

the pattern

Or did you take the easy way out

and just click

Page 13: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

13To put your footer here go to View > Header and Footer

So making full use of CAST

Page 14: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

14To put your footer here go to View > Header and Footer

Interact and read the text as well

Instructions

Instructions and statistics

Important points are in white

Page 15: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

15To put your footer here go to View > Header and Footer

Using Excel effectively

• Dot plots are not on Excel’s menus

• Dot plots are not in Excel’s help

• But you decided to do dot plots in Excel!• You therefore need to understand them better• So you can construct them yourself• And this understanding is good anyway• And helps with effective data analysis

• It is an example• Of you controlling the software• And not being limited by it

• That applies to all software

Page 16: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

16To put your footer here go to View > Header and Footer

Jittered dot plots in CAST and Excel

CAST

EXCEL

Why are the vertical heights different in the 2 cases?

Do you ALL know?

Page 17: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

17To put your footer here go to View > Header and Footer

Excel for analysis and training

• Excel is not designed as a training resource• Unlike CAST – that is all CAST is for

• Excel is to support • data organisation• and analysis

• But here we have used it also for training• With dot plots• And stem and leaf plots• Neither of which are in the Excel menus

Page 18: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

18To put your footer here go to View > Header and Footer

Summary• Dot plots and stem & leaf plots give simple tools

• to look at the actual data in a simple and concise way

• It is important to look at the data itself • before starting on the actual analysis • so any patterns or oddities can be identified • and necessary steps taken to deal with them

• When dealing with large sets of data, computers are needed to do the exploration;

• However the importance of this work • should be stressed right at the data entry stage • and could even become part of the data checking

procedures

Page 19: SADC Course in Statistics Exploratory Data Analysis for single variables Module B2 Session 12

19To put your footer here go to View > Header and Footer

The next session will extend and apply the tools from this session to real data