Université de Nice Sophia-Antipolis: Accueil
Notes for courses by
D a v i d   C r o o k a l l

AreasCourses / Presentations / Job interviews / Teamwork / X-culture / Searching / Research / Publication / Communication / Projects / Meetings / Studying
CoursesInfoCom (ACL) L-3 • IUP Com SEDI L3 • IUP Job interviews • IUP Teamwork • IUP x-cult • MBFI • Telecom

Stats č • Up • Research cycle • Stats • Thesis writing • About research • Data_sources_Stats • Resources •

Research cycle
Thesis writing
About research

Simulation & Gaming:
An Interdisciplinary Journal



Stats • Up •
• Glossary •
• Topics •
• What stat? •
• Nonparametric •
• Links •
• SPSS •

A few of the many stats resources on the web: Data and methods

On-line books

Introductory Statistics: Concepts, Models, and Applications (by D. Stockburger)

A New View of Statistics: On-Line Stats Book (by W. Hopkins)

HyperStat: On-Line Stats Book (by D. Lane)

The Little Handbook of Statistical Practice (by G. Dellal)

From  www.ku.edu/%7Ecoms/virtual_assistant/vsa/index.htm  &  http://www.ukans.edu/cwis/units/coms2/vsa/index.htm

The Virtual Statistical Assistant 1.0
University of Kansas

Welcome to the University of Kansas Department of Communication Studies Virtual Statistical Assistant (VSA). The VSA employs a decision tree technique designed to help determine which statistical techniques are most appropriate for a given research design. Just follow the prompts and the VSA will guide you through the process of selecting the appropriate statistic. If you are interested in more general questions about social science research you might explore the Virtual Research Assistant.

If you are not satisfied with the decision tree model utilized in this version of the Virtual Statistical Assistant you might try the grid selection model which is provided as an alternative method in this virtual assistant.

Determine The Level of Measurement

The first step in using the virtual statistical assistant involves determining the level of measurement (data type) collected on the dependent (outcome) variable. Choose the level of your data as best described by the following choices:


Nominal level data are categorical. That is the data are expressed in terms of categories. The data cannot be arranged in any order with respect to one another. Some examples of nominal level data are marital status (never married, divorced, separated, widowed and married), and religious or political affiliations.


Categories are ordered, but differences cannot be determined or they are meaningless. Ordinal level data are also categorical but the categories can be ordered with respect to one another. The ordinal level of measurement involves data that may be arranged in some order, but differences in data values either cannot be determined or are not important. One example of ordinal data are socioeconomic status (lower middle, middle, or upper middle classes). In this example lower middle class denotes a smaller income than in the other two classes but the amount of the difference is arbitrary. Other examples include car sizes subcompact, compact, midsize, luxury), restaurant/hotel ratings (one star, two star, five star, etc), or in survey questionnaires where the respondent is asked to rank order selected choices.


Interval and ratio level data have categories that are ordered and meaningful differences can be determined. Interval data differ from ratio data inasmuch as there is no inherent absolute zero point. With ratio data there are an absolute zero (i.e. ratio data is always a subset of interval data, but there is a value of absolute zero). For example interval data include temperature, time (Gregorian calendar measurements) or in survey questionnaires where the respondent is asked to rate items along a continuum with equal and known numerical intervals (as with a Likert-type scales). Examples of ratio data include age, weight, height, or distance. Now that you have decided you have Interval/Ratio data your next step is to determine whether you are measuring a relationship or a difference.

If you still are not certain how to start this selection process you should look at the grid selection model provided by the Virtual Statistical Assistant.

Before selecting an appropriate statistic we recommend you consult the following sources:

PROPHET StatGuide (An online guide to statitics and their use.)

A Practical Guide to the Use of Selected Multivariate Statistics (A Great Source!).

See also:

SurfStat Australia (Another valuable statistics site with interesting links.)

Page URL: http://www.ukans.edu/cwis/units/coms2/vsa/index.html

The VSA was constructed by students in the Department of Communication Studies. Special thanks to Clark, Tina & Mei-Chen. Please send feedback and comments to VSA staff. You might also see the Selecting Statistics site at Cornell University - a virtually identical site (developed independently we can assure you).

From  http://www.earthresearch.com/data-importance.shtml

"Down to Earth" Research Advice

Written for you by Dr J. Mark Tippett, www.earthresearch.com

The Importance of Data

The need for data

Most research projects need data in order to answer a proposed research problem. The data that need to be acquired, and the sources of such data, must be identified as a matter of utmost importance. No amount or depth of subsequent data analysis can make up for an original lack of data quantity or quality.

Research problems and objectives (or hypotheses) need to be very carefully constructed and clearly defined, as they dictate the data that need to be obtained and analyzed in order to successfully address the objectives themselves. In addition, the quantity of data, their qualities, and how they are sampled and measured, have implications for the choice and effectiveness of the data analysis techniques used in subsequent analysis.

Fundamental questions

Fundamental questions to be asked (and hopefully answered) with respect to the proposed research and data include:

What data are needed?

What data need to be measured or obtained? What are the required characteristics of the data in terms of their quantities and qualities?

Do the data already exist and can they be obtained?

If so, what are the sources of the data? How were the data measured? What are the characteristics of the data in terms of their type, quality, resolution, precision, accuracy, and coverage? Is the quantity of data sufficient? Are their characteristics suited to, and sufficient for, the study? How will you actually assess their suitability?

If the data do not exist, what data need to be generated?

What data characteristics are required in terms of data type, quality, quanitity, resolution, precision, accuracy, and coverage, in order to properly address the research objectives? What variables will be measured? How will meaurements be made? What sampling scheme will be employed, and why? What logistical problems (e.g., accessibility) need to be considered? At what scale(s) will measurements be made? How will you ensure that you are measuring what you think you are measuring (a tricky one!)?

What implications are there for the subsequent analysis?

How does sample size constrain the effectiveness (e.g., power) of statistical tests? Are replicate observations needed? Is there a spatial dimension to your data, and if so have you worked out what the distance between your samples should be? Have you over-sampled or under-sampled, and can this be remedied beforehand? Are the data "representative" and how do you know? Are the data "random" or "stratified" or "nested" and does this matter? Does the type of data - ratio scale, interval, ordinal, discrete, nominal, closed, directional - have implications for data analysis (Yes!)?


  • Most research requires data and data analysis.
  • Data acquisition is of utmost importance and considerable effort should be made to obtain or generate good data.
  • Good data are data whose characteristics enable the research objectives to be met.
  • Data of poor quality or undesirably low quantity will lead to unsatisfactory data analysis and vague results.
  • The characteristics of the data, particularly their type, quantity, and how they were sampled, constrain the choice of data analysis techniques able to be used on the data.
  • Data analysis can only be as good as the original data allow.

If you are a doctoral student or funded researcher in need of help with your research proposal or data analysis, I am qualified to help you with the statistical aspects of your study and I am prepared to work to your schedule

Statistics Solutions exists to help dissertation and thesis students graduate.
Statistics Solutions, Inc. is a statistics research company of doctorate-level statisticians dedicated to helping you finish your dissertation or thesis. Through our personal experience of completing a dissertation, we know what it takes to design a study and analyze the data.




 • Research cycle • Stats • Thesis writing • About research • Data_sources_Stats • Resources •

RSS Recent issues of Simulation & Gaming: An Interdisciplinary Journal
Peace and survival of life on Earth as we know it are threatened by human activities that lack a commitment to humanitarian values.  Destruction of nature and natural resources results from ignorance, greed, and a lack of respect for the Earth's living things... .  It is not difficult to forgive destruction in the past, which resulted from ignorance.  Today, however, we have access to more information, and it is essential that we re-examine ethically what we have inherited, what we are responsible for, and what we will pass on to coming generations.  Clearly this is a pivotal generation... .  Our marvels of science and technology are matched if not outweighed by many current tragedies, including human starvation in some parts of the world, and extinction of other life forms... .  We have the capability and responsibility.  We must act before it is too late.  Tenzin Gyatso the fourteenth Dalai Lama.