STATISTICAL TOOLS AND SOFTWARES BY GHANA STATISTICAL SERVICE

STATISTICAL TOOLS AND SOFTWARES BY GHANA STATISTICAL SERVICE OCTOBER, 2018 1

PRESENTATION OUTLINE 2

INTRODUCTION (1/3) • Statistical Analysis Tools are required for a thorough and scientifically valid analysis of survey results. There are several choices available for the researcher to choose from – ranging from the simple tools available with all survey packages that calculate percentages and totals – to the very advanced tools requiring a graduate education to learn and use. • Basic analyses can be easily computed, while more advanced methods require a solid understanding of advanced statistics as well as specialized computer software. 3

DESCRIPTIVE ANALYSIS (1/2) • Descriptive analysis uses specific tools to describe data. • These are relatively simple calculations that give a basic picture of what the data looks like overall. • Descriptive tools include: frequency, percentages and measures of central tendency. 4

DESCRIPTIVE ANALYSIS (2/2) • Frequency tells how many times something has occurred in a data set. • Percentages are calculations that show a proportion. • Measures of central tendency are represented by the mean, median and mode. • These tools describe the central point (median), the most common (mode) or the average (mean) for a specific variable. 5

MODERATE ANALYSIS (1/2) • Moderate statistical analysis tools look at the relationships between variables: -- • what the nature of these relationships are and if they are significant. • These include correlation and regression. 6

MODERATE ANALYSIS (2/2) • A correlation describes the relationship between two variables as well as the direction and strength of that relationship. • Regression variable. can show if a variable predicts another • Like correlation, however, regression does not show causation 7

ADVANCED ANALYSIS (1/3) Advanced analyses include calculations of variance. • These can help a researcher see what variety exists in the data, as well as positive outcomes in the research. • In order to calculate variance, a researcher must use the standard deviation. • A standard deviation measures the degree that an individual value varies from the mean or average. • Once the standard deviation is known, analysis of variance can be conducted. 8

• An ADVANCED ANALYSIS (2/3) analysis of variance or ANOVA is used to compare the difference in the means or averages of variable groups. • This will show if an outcome from one group is statistically different from the outcome for another group. • An Analysis of Covariance, or ANACOVA, is a tool that can be used for experimental research designs. • ANACOVA will tell the researcher the variance between pre- and post-test data. 9

STATISTICAL SOFTWARES 10

INTRODUCTION (1/2) • We carry out research to test hypotheses, and we do that by getting hold of data. • Hopefully, if our experiments are planned and executed correctly, we can get hold of good data that can tell us something unique about the world. • . While the first part of any experiment – the planning and execution – is critically important, it is only half the battle. • How the data is treated is just as important, and analyzing good data in the 11 right way can lead to groundbreaking findings and insights.

INTRODUCTION (2/2) • Data analysis is often seen as the most scary aspect of completing research, but it doesn’t have to be that way. • While you will need to understand what to do with the data, and how to interpret the results, software that is designed for statistical analysis can make this process as smooth and as easy as possible. • A great number of tools are available to carry out statistical analysis of data, and below (in no particular order are afew packages suitable for human behavior research. 12

STATISTICAL SOFTWARES (SPSS) • SPSS, (Statistical Package for the Social Sciences) is perhaps the most widely used statistics software package within human behavior research. • A general-purpose statistical package widely used in academic research for editing, analyzing and presenting numerical data. • SPSS offers the ability to easily compile descriptive statistics, parametric and non-parametric analyses, as well as graphical depictions of results through the Graphical User Interface (GUI). • It also includes the option to create scripts to automate analysis, or to carry out more advanced statistical processing 13

STATISTICAL SOFTWARE (R) (R Foundation for Statistical Computing) • • • R is a free statistical software package that is widely used across both human behavior research and in other fields. Toolboxes (essentially plugins) are available for a great range of applications, which can simplify various aspects of data processing. While R is a very powerful software, it also has a steep learning curve, requiring a certain degree of coding. It does however come with an active community engaged in building and improving R and the associated plugins, which ensures that help is never too far away. R provides a wide variety of statistical (linear and nonlinear modelling, classical statistical tests, time-series analysis, classification, clustering, etc. ) and graphical techniques, and is highly extensible. R is freely available online. 14

MATLAB (The Mathworks) • Mat. Lab is an analytical platform and programming language that is widely used by engineers and scientists. • As with R, the learning path is steep, and you will be required to create your own code at some point. • A plentiful amount of toolboxes are also available to help answer your research questions • While Mat. Lab can be difficult to use for novices, it offers a massive amount 15 of flexibility in terms of what you want to do – as long as you can code it (or at least operate the toolbox you require).

Microsoft Excel • While not a cutting-edge solution for statistical analysis, MS Excel does offer a wide variety of tools for data visualization and simple statistics. • It’s simple to generate summary metrics and customizable graphics and figures, making it a usable tool for many who want to see the basics of their data. • As many individuals and companies both own and know how to use Excel, it also makes it an accessible option for those looking to get started with statistics. 16

SAS (Statistical Analysis Software) • SAS is a statistical analysis platform that offers options to use either the Graphical User Interface (GUI), or to create scripts for more advanced analyses. • It is a premium solution that is widely used in business, healthcare, and human behavior research alike. • It’s possible to carry out advanced analyses and produce publication-worthy graphs and charts, although the coding can also be a difficult adjustment for those not used to this approach. 17

Minitab • The Minitab software offers a range of both basic and fairly advanced statistical tools for data analysis. • Similar to Graph. Pad Prism, commands can be executed through both the Graphical User Interface (GUI) and scripted commands, making it accessible to novices as well as users looking to carry out more complex analyses. 18

Graph. Pad Prism • Graph. Pad Prism is premium software primarily used within statistics related to biology, but offers a range of capabilities that can be used across various fields. • Similar to SPSS, scripting options are available to automate analyses, or carry out more complex statistical calculations, but the majority of the work can be completed through the Graphical User Interface (GUI). 19

Stata • A powerful and flexible general-purpose statistical software package used in research, among others in the fields of economics, sociology, political science. • It's capabilities include data management, statistical analysis, graphics, simulations, regression, and custom programming. 20

21

THANK YOU 22