Using Statistics in Management Research

1
Using Statistics in Management Research

• Dr. Barbara L. Marcolin
• MIS Area
• Faculty of Management

2
Welcome

• Email addresses
• Course outline
• Web site
• Labs
• Introductions
• Remainder of first day
• Introduction to key statistical terms
• Nature of quantitative data
• Running SPSS data files and syntax files

3
Multivariate Definition

• More than two variables (i.e., Multivariate)
• Assumed to have a multivariate normal data
  distribution
• All variables are extremely interrelated such
  that it is difficult to separate them for
  interpretation.

4
Measurement

• Scale
• Non-metric (qualitative)
• Nominal ( can assign to categories) or Ordinal
  ( can be ordered or ranked)
• Metric (quantitative)
• Interval ( constant unit of measurement without
  an arbitrary zero point) or Ratio ( constant
  unit of measurement with an absolute zero point)
• Error in measurement
• Reliability
• Degree to which the observed variable measures
  the true value and is error free
• Opposite of measurement error
• Validity
• degree to which a measure accurately represents
  what it is supposed to measure

5
Significance

• When inferring from a sample to a population, the
  researcher must specify an acceptable level of
  error.
• Alpha or Type I error is the chance of a false
  positive OR falsely concluding that there is a
  difference when there isnt one.
• Beta or Type II error is the inability to reject
  the null hypothesis (I.e., not being able to
  detect a difference when you should).
• Power (1-beta) is the probability of being able
  to reject the null hypothesis.
• Power is a function of effect size, alpha, sample
  size and measurement error.

6
Key Terms

• Alpha (?) - Type I Error plt.05
• Beta (?) - Type II Error
• Power
• Practical significance
• Statistical significance
• Dependence technique
• Dependent variable

7
Key Terms (continued)

• Interdependence technique
• Independent variable
• Hypothesis Testing
• Dummy variable
• Effect size
• Indicator
• Non-metric data
• Metric data

8
Key Terms (continued)

• Multicollinearity
• Summated Scales
• Reliability
• Treatment
• Anova
• Multivariate analysis

9
Types of Univariate Techniques

• Parametric
• t
• variance test (?2)
• T-tests
• unparied (t-test groups )
• paried (t-tests pairs )
• F-tests
• Z test

• Non-Parametric
• Mann-whitney
• Sign test
• Wilcoxon Signed Rank
• Wilcoxon Rank Sum
• Kruskal-Wallis

10
Bivariate Techniques

• Pearson product-moment correlation x continuous,
  y continuous.
• Spearman r correlation (non-parametric)
• Kendalls tau correlation (non-parametric when
  there are ties)
• Chi-square Goodness of fit test
• Contingency tables and Fishers Exact test
• Point biserial correlation x dichotomous, y
  continuous.
• Phi coefficient x dichotomous, y dichotomous

11
Types of Multivariate Techniques

• Multiple Regression
• Analysis of Variance (ANOVA)
• SPSS oneway command
• ANCOVA
• Two-group discriminant analysis
• Logit (Multiple Frequency analysis)
• two group logistic regression

• Canonical Correlation
• Multiple Analysis of Variance (MANOVA) (SPSS GLM
  command)
• MANCOVA
• Profile Analysis
• Multiple Discriminant Analysis
• Factor Analysis
• Principal Components Analysis
• Structural Equation Modelling - LISREL, EQS,
  AMOS, SEPATH, MX
• Structural Equation Modelling - PLS
• Conjoint Analysis
• Multidimensional Scaling
• Cluster Analysis
• Correspondence Analysis