Microsoft Office Tutorials and References

In Depth Information

4. Use data analysis to generate further questions of interest. In the case of the

teen’s data, we made no distinction between male and female teens, or the actual

ages of the teens. It is logical to believe that a 13 year old female web visitor

may behave quite differently than a 19 year old male. This data may be available

for analysis and it may be of critical importance for understanding behavior.

Often our data is in qualitative form rather than quantitative, or is a combination

of both. In the next chapter, we perform similar analyses on qualitative data. It is

important to value both data types equally, because they can both serve our goal of

gaining insight. In some cases, we will see similar techniques applied to both types

of data, but in others, the techniques will be quite different. Developing good skills

for both types of analyses is important for anyone performing data analysis.

Key Terms

Add-in

Series

Treatment

Time Series Data

Cross-sectional Data

Cyclicality

Seasonality

Leading

Trend

Linear Trend

E-tailer

Page-views

Frequency Distribution

Central Tendency

Variation

Descriptive Statistics

Mean

Standard Deviation

Population

Range

Median

Mode

Standard Error

Sample Variance

Kurtosis

Skewness

Systematic Behavior

Linear Regression

Dependent Variable

Independent Variable

Simple Linear Regression

Beta

Alpha

R-square

Residuals

Signiﬁcance F

Covariance

Correlation

Perfectly Positively Correlated

Perfectly Negatively Correlated

Winters’ 3-factor Exponential

Smoothing

Exponential Smoothing

Level of Conﬁdence

t-Test

t-Test: Paired Two Sample for Mean

Type 1 Error

t-Stat

Critical Value

Test of Hypothesis

Null Hypothesis

One-tail Test

Two-tail Test

ANOVA

Search JabSto ::

Custom Search