the problem in word and the data and directions  in attachments 

Problem #3 Modification and Recoding of Data

Part 1:

1. Data set is Group

2. Variables of interest are Year and Age

Directions:

Determine the average age of students, based on their class ranking of either Lower or Upper class standing. This requires the creation of a new variable. Examine the variable Year to create new variable based on current class standing. Be sure to give the new variable groupings a relative name. To compare the average age, run the “Means” test, Analyze>Compare Means>Means to obtain the statistics.

1. Recode class year ( variable is Year) into Lower Class and Upper Class.

a. Lower Class is Freshman and Sophomore

b. Upper Class is Junior and Senior

2. What is the average age of Lower class students?

3. What is the average age of Upper class students?

Part 2:

1. Data set is Census90

2. Variable of interest is Age

Directions:

1. Recode age into Young and Old

a. Young is any age less than 40

b. Old is any age 40 and above

Follow the procedures in Part One. In addition, run Descriptive statistics for Skewness, Kurtosis, Histogram with Normal Curve, and Normality test on the overall variable, Age.

2. a. How many Young people are there?

b. How many Old people are there?

3. a. What is the average age of Young people?

b. What is the average age of Old people?

Transforming data to fit the application

Introduction to Data Modification
Adapting data to fit the application

Recode existing varaibles
Compute a new variable
Selecting cases from a data file
Splitting the data file into groups
Selecting cases and splitting files are not data modification, but are included as steps preceding descriptive and inferential statistical analysis procedures.

Recode Variables
Variables are recoded for two reasons

1. To modify certain values of a variable
2. To group certain values into a single value.
When recoding, two options are available
1. Recode into same variable
2. Recode into different variables.

Recoding into Same Variables
When using this option, the variable keeps the same name and labels, with the new values in the same column.
Original data is lost
No way to undo the action
Not advisable method

This method retains the original variables
Values are added to the data file under a new variable name in the first empty column of the file.

Transform > Recode into different variables
Recode Variables

Recode Variables Create new variable

Recode Variables Create new value

Select Variable View and provide names for the two values of the “youngold” variable.
1 = young
2 = old
New variable is now ready to use.
Recode Variables Name new value

SPSS provides a way to create new variable from a combination of other existing variables.
For example, if we have data measuring computer skills on several different indicators. We could add the scores similar to an interval scale to achieve a ranking as in a Likert scale.
Transform > Compute Variable
Create New Variable

Create New Variable

Select Cases
Used when analysis is run only on a selected subset of cases in the data file.
Data > Select Cases

Select Cases

Select Cases Filter remains on (lower right) until removed

Split Files
Used to separate the file into groups, such as male and females.
Data > Split File

Split Files
Analysis proceeds now based on the two separate groups.
Go to Data > Split File > Reset
Recombines the data file into one group