Basic Level Module B2- Handling Data

Document Sample
Basic Level Module B2- Handling Data Powered By Docstoc
					                                                              Basic Level Module B2

Module B2: Handling Data


This module provides an overview of what happens to data once they have been collected.
It includes the organisation of simple datasets for analysis and describes the basic ideas of
producing good tables and graphs. Throughout this module students will be introduced to
principles that will help making efficient use of spreadsheets for statistical work; they will
be introduced to exploratory data analysis techniques and some summary statistics; they
will develop familiarity with the electronic book “CAST for SADC” (Computer Assisted
Statistics Teaching); and they will learn basic principles for data quality assessment and


Management of Data

Successful students will

       Be able to structure a simple data set in a format that is suitable for statistical
        processing using a spreadsheet program.
       Be able to code, sort, filter and rearrange data from a list in a spreadsheet.
       Be aware of the aspects that are important to maximise data quality
       Be able to archive a data set ensuring that all relevant information is included.

Statistical techniques

Successful students will

       Be able to correctly explain basic concepts such as population, sample and
        summary statistics.
       Be able to interpret the meaning of some basic summary statistics including:
        frequencies and percentages.


Successful students will

SADC Course in Statistics                                                   Module B2 – Page 1
                                                         Basic Level Module B2

      Be able to summarise a statistical report written for a general audience.
      Be able to assess whether a graph has been constructed following good practice.
      Be able to improve a poorly constructed graph
      Be able to assess whether a table has been constructed following good practice.
      Be able to improve a poorly constructed table

IT techniques

      Be able to use CAST as an aid to learning statistics
      Be able to use Excel/SSC-Stat to produce basic statistical summaries and tables.
      Be able to use Excel/SSC-Stat to produce simple graphs.


Computing – Some experience of using a spreadsheet (Excel) and word processor.

SADC Course in Statistics                                             Module B2 – Page 2
                                                            Basic Level Module B2


Session 1. From the data to the report

Examples of the management steps followed in statistical studies. Three different cases
will be considered: a small study, probably from a student project, one medium size study
probably related to a snap-shot survey, and one large study likely to be drawn from an
integrated household survey.

Session 2. CAST – an electronic statistics textbook

Demonstration of CAST and first use of Book

Session 3. Statistical concepts

Examples used to review statistical concepts and terminology, including population,
sample, summary statistics, descriptive and inferential statistics, types of data – numerical
and categorical, types of presentation – exploratory and presentation graphs, tables.

Sessions 4 & 5. Producing good tables in Excel

The use of tables in summarising data is introduced and practiced. In the first of these two
sessions, the frequencies are provided or calculated by hand. They are then transformed
into proportions and percentages.

The second session uses Excel’s pivot table facility to calculate the frequencies (counts),
which are then edited to produce an appropriate and readable summary of the data.

These 2 sessions are designed to give students the (hand and computing) skills to produce
and manipulate simple frequency tables. Deciding what is a good or poor table is in a later

Session 6 & 7. Producing good graphs and charts using Excel

Session 6 describes the use of bar charts and pie charts from data that are summarised in a
form that is ready for the chart to be constructed. The use of Excel’s graphics wizard is

SADC Course in Statistics                                                 Module B2 – Page 3
                                                           Basic Level Module B2

Session 7 is the parallel of Session 5, where Pivot tables were introduced. This is now
taken a step further, so both the table and (pivot) graph are constructed.

Session 8. Presenting good tables and graphs

Examples of reports, concentrating on counts, frequencies and percentages in tables and
graphs (bar charts and pie charts). This session will include examples of tables and graphs
that follow good practice and examples of tables and graphs that can be improved.

Session 9 & 10. Organising data in a spreadsheet

This session will discuss basic principles of data organisation in a spreadsheet such as the
use of lists, names and comments. It will include discussions on the uses and limitations of
a spreadsheet for data entry. It will highlight strengths and weaknesses of the use of
spreadsheets for statistical work.

Session 11. Excel for statistics

Introduction to a simple add-in (SSC-Stat) that encourages good statistical practice for
spreadsheet work. Basic data manipulation: stacking and un-stacking columns.

The session makes use of a tutorial introducing SSC-Stat. This continues the provision of
resources to help students to learn for themselves.

Session 12. Exploratory data analysis (EDA) for single variables

This session introduces the idea of exploratory data analysis as a stage before the formal
analysis is undertaken. It concentrates on two simple tools, namely dot plots and stem and
leaf plots.

The session also considers how students can ensure they use resources effectively. This is
illustrated through discussions on the use of Excel and CAST.

Session 13. EDA in practice

The last session introduced some simple tools for data exploration. This session uses those
and other tools to look at exploratory data analysis in general.

SADC Course in Statistics                                               Module B2 – Page 4
                                                             Basic Level Module B2

Data from Tanzania and Swaziland illustrate the importance of EDA, before and at the
start of the data analysis.

Session 14. Objectives and analysis

This session emphasises the importance of having clear objectives. How clear statements of
the objectives lead to appropriate statistical reports. How you should be able to plan you
analysis based on the objective even before collecting the data. It also looks at it the other
way on by providing graphs and tables for which the objectives need to be identified.

Session 15.Analysing numeric variables

Adding codes for categorical data and recoding quantitative data. Students learn some new
excel skills for analysing numeric variables and expand their table making skills. These new
techniques are then applied to the sunshine data to further hone their new skills.

Sessions 16 & 17. An analysis in practice

These two sessions centre on a Group project which use the methods and skills acquired
throughout this module.

The project work is to be done in groups of 4 and will take up most of the day. As well as
providing an opportunity to practice and consolidate the learning done so far on exploring
and describing data, these sessions introduce participants to the importance of “soft” skills
such as communication skills, working as part of a team and working to a schedule.

Sessions 18 and 19. Reading a statistical report

What is a good report statistical report? Targeting the audience, structuring the report,
good practice in report writing, revisions, proof-reading and publication. Referencing
sources and plagiarism.

These sessions require students to read a statistical report that puts into practice the skills
and knowledge developed throughout this module. The report will include simple data
analyses to produce summary statistics, frequencies and percentages, and the corresponding

SADC Course in Statistics                                                   Module B2 – Page 5
                                                           Basic Level Module B2

Session 20. Presenting a report

In this session students give the PowerPoint presentations they put together in session 16
& 17. As each group presents the observers complete evaluation forms to be provide praise
and constructive criticisms in the class discussion at the end of the session. Self evaluation
is also part of this session.

SADC Course in Statistics                                                Module B2 – Page 6