Quantitative Data Analysis or Introduction to Statistics

1
Quantitative Data AnalysisorIntroduction to
Statistics

• Need to know
• What you want to know --
• Relationship? Prediction? Causation?
  Description? Everything?!
• Appropriate statistical procedures
• Depends on what you want to know.

2
Common Statistical Procedures
Relationship between 2 variables
Correlation
Relationship among more than 2 variables
Factorial analysis
Prediction
Regression Chi-square
Description of pheno- mena or variables
Descriptive statistics (frequency, mean, etc.)
Statistical control
Standardization
3
Correlation

• Mathematical ex-pression of the nature of a
  relation-ship between 2 variables
• Expressed in terms of a coefficient
• Relationship may be strong or weak, direct or
  inverse

4
ANOVA, MANOVA

• MANOVA allows for the study of two or more
  rela-ted dependent variables
• Effect of each IV without consideration of others
  is the main effect
• Effects of two or more IVs combined is the
  interaction effect

• Used to determine whe-ther obtained differences
  between means of two or more groups are due to
  chance
• The larger the value of F, the more likely the
  result is statistically significant
• Used when more than one independent variable is
  investigated

5
Regression

• Regression analysis techniques allow the
  researcher to predict the behavior or some
  variables based on the knowledge of the behavior
  of others
• Regression also refers to the tendency of scores
  to move toward the mean of the population in
  repeated tests

6
Chi-Square

• May be used to deter-mine the probability that an
  observed relationship is statistically
  significant
• On a single independent variable, tests goodness
  of fit, or How well do sample values
  corre-spond to hypothesized population values?

• On two independent variables, tests
  indepen-dence, or Are the values of one
  variable related to, or dependent on, the values
  of the other independent variable?
• Assumes random selec-tion of samples,
  indepen-dent observations, sufficient sample
  sizes

7
Descriptive Statistics

• Often used to organize or display data
• Can be used to summarize important
  characteristics of a set of numerical data
• Describes the perform-ance variability of
  sample scores graphically

8
Standardization

• Conversion of raw scores on different measures to
  a single scale for the purpose of comparison
• Also describes a single distribution of z scores
  considered normal, where the mean is 0 the
  standard deviation is 1

• Two common formats are z scores and T scores
• IQ scores are an exam-ple of standard scores
  normally distributed, with a mean of 100 and a
  standard deviation of 15 (Wechsler)