You are on page 1of 20

Introduction

Statistics is the art and science of collection, presentation, analysis, and interpretation of data. Statistics are numerical facts that are systematically collected or analyzed. Think of it as sheep, which can be both singular and plural.

Condenses large quantities of information into a few simple figures or statements Aids in decision-making Gives basis for comparison Justifies a claim or assertion Helps in finding a relationship Predicts future outcome

Sports Research Health Predictions

Statistics is for YOU!!!

Descriptive and Inferential Statistics


Descriptive statistics consists of the collection,

organization, summarization, and presentation of data. Inferential statistics, on the other hand, uses probability. It also generalizes from samples to populations, performs hypothesis testing, and determines relationships among variables.

Population (Totality) and Sample (Sub-group) Quantitative (Numerical) and Qualitative (Categorized) Discrete (Countable) and Continuous Variables Parameter (from the population) and Statistic (from the sample)

If, in a US study, it is found that lightning hits more men (376) than women (63) how might this information be used by an insurance company?

Imagine that there is a study that seeks to know how many men want to know the gender of their wives unborn children. Lets say that 25% of the men want to know, and the remaining 75% do not want to. How may we define the population? How may we define the sample?

Nominal
Mutually exclusive (non-

Ordinal

Classifies data into

overlapping) Exhaustive No order or ranking can be imposed Best and easiest examples: Gender and Course

categories that can be ranked Precise differences do not exist between the ranks though. Examples: Letter grades, attitude scales, peoples builds

Interval
Ranks data Precise differences exist No meaningful zero Examples: IQ Tests,

Ratio

Has all the properties of

Celsius and Fahrenheit temperature scales

the interval level, but also has a meaningful zero (where zero signifies total absence). True ratios exist between different units of measure Weight, Length, and Income

Intelligence Quotient Lapsed time Eye color Course Tournament Ranking UPCAT score Nationality Height Gold medals won ZIP code

Validity
The extent to which a

test measures what we actually want to measure The degree to which they accomplish the purpose for which they are being used.

Reliability
The accuracy and

precision of a measurement procedure The extent to which an experiment, test, or any measuring procedure yields the same result on repeated trials.

Availability
Source Series
Cross Internal External Primary Secondary

Sectional Cohort Panel

Census/Survey
Personal Interview Telephone Interview

Self-administered Questionnaire

Experiment Naturalistic Observation


No manipulation of variables is done

Probability Sampling
Simple Random

Non-Probability Sampling
Convenience Purposive Quota Snowball

Sampling Systematic Sampling Stratified Sampling Cluster Sampling Multi-Stage Sampling

This is the reduction of a wide variety of idiosyncratic information to a more limited set of attributes composing a variable. Coding must be done in somewhat more detail than what you plan to use in the analysis. Keep in mind: code categories must be exhaustive and mutually exclusive.

This is a document used as the primary guide in the coding process. It helps you locate variables and interpret codes during the analysis stage. There are certain requisites:

Variables should be identified by an abbreviation. The full definition of the variable should be in the

codebook. The exact wordings of questions must be contained.

This is the end product of the coding process: the conversion of data items into numerical codes. We can use SPSS and MS Excel. Later you will see a demonstration of encoding in both programs.

Dirty data will almost always produce dirty findings. We can clean data using two methods:
Possible code cleaning SPSS: Variable definition MS Excel: Validation Contingency cleaning Logic and common sense. For example, if you see height listed as 220cm, you may want to go back to the originals.

You might also like