Anda di halaman 1dari 17

INTRODUCTION TO

SPSS
1

SPSS

Statistical Package for Social Sciences


In

1968, Norman H. Nie, C. Hadlai (Tex) Hull


and Dale H. Bent, three young men from
disparate professional backgrounds,
developed a software system based on the
idea of using statistics to turn raw data into
information
The initial work on SPSS was done at Stanford
University with the intention to make it
available only for local consumption and not
international distribution
6

Nie, a social scientist and Stanford


doctoral candidate, represented the
target audience and set the
requirements
Bent, a Stanford University doctoral
candidate in operations research, had
the analysis expertise and designed
the SPSS system file structure
Hull, who had recently graduated from
Stanford with a master of business
administration degree, programmed.

McGraw-Hill

published the first SPSS


user's manual in 1970
Nie, Bent, and Hull received a royalty
from sales of the manual but nothing
from distribution of the program. In
Nie's words, "It was like Gillette selling
razors at cost and getting its profits
from the blades."

Windows In SPSS
Data

View
Variable View
Output

DATA VIEW window

10

Variable View Window

11

Output window

12

Name
Age
Sex

Male

Female

Rural

Urban

Previous Degree

BA

Bcom

Bsc

Current Degree

MBA(Reg)

MBIT

MBA(EXE)

Current Residential Status

Boarder

Non Boarder

Fathers Education

Illiterate

Literate

Primary

Middle

Matric

Inter

Graduation

Master

Blood Group
City
Family Background
Family Income
Others

Others
Mothers Education

Illiterate

Literate

Primary

Middle

Matric

Inter

Graduation

Master

Others
Fathers Occupation

Govt. job

Private Job

Business

Others

Mothers Occupation

Govt. job

Private Job

Business

House Job

Others

Graduation Marks
Marks in FA
Marks in HRM
13

Marks in Math & Stat


Facilities

InterNet

Cell

Conv.

Measurement Scales
Data can be classified according to
levels of measurement. The level of
measurement of the data often
dictates the calculations that can be
done to summarize and present the
data. It will also determine the
statistical tests that should be
performed.
14

Measurement Scales
Types of
Measurement
Scales

Nominal

Ordinal

Interval

Ratio

Data may only


be classified

Data are
ranked

Meaningful
Difference
Between values

Meaningful Zero
point and Ratio
Between values

Eye colour,
Religion,
Sex, etc.

Cricket teams
standings in
ICC ranking
Students grades, etc

Temperature,
Shoe Size,
IQ Scores

Bank Balance,
Weight,
Height, etc.
15

Defining variables
Variable Name:Each measured
character must be
identified with a
unique NAME
Space & hyphen is not
allowed in variable
name
First character must
be alphabet
Under score is allowed

EXAMPLE:
REGD NO

(Wrong)
REGD-NO

(Wrong)
2005_REGD

(Wrong)
_REGD

(Wrong)
RegistrationNo

16

Defining the attributes of


variables

Type

(or data type). Data can be of several


types, including numeric, text, currency, and
others
TYPE

EXAMPLE

Numeric

1000.05

Comma

1,000.005

Scientific

1 * e3
(the number means 1 multiplied by
10 raised to the power 3, i.e.
(1)*(103)

Dollar

$1,000.00

String

Ali shah
17

Defining the attributes of


variables

Variable label.
Defining a label for a
variable makes
output easier to read
but does not have
any effect on the
actual analysis For
example, the label
"Family Identification
Number" is easier to
understand than the
name of the variable,
fam_id.

18

Defining the attributes of


variables
Column format:Column format:allows you to choose
the width of the
column as displayed
on screen and to
choose how the text
is aligned in the
column (left, center,
or right aligned).

19

Defining the attributes of


variables

Value labels:
Enable the use of labels
instead of values for
specific values of a
variable, thereby
improving the quality of
output. For example,
for the variable gender,
the labels "Male" and
"Female" are easier to
understand than "0" or
"1.

20

DATA

21

Anda mungkin juga menyukai