Professional Documents
Culture Documents
An Introduction
SPSS stands for
Statistical
Package for the
Social
Sciences
Variable Classification
• Nominal variable
Ordinal variable
examples possible raw data
names Juan, Dexter, Agnes, Marife, …
Numeric variable
Examples
baranggay
birth name
possible
Tetuan,
order first,
raw data
Labuan,
second, Zone
third, III,…
last, San Jose, …
monthly
household salary
examples
number possible
001, raw data
below002, 003,
Php1000, 004, …
Php1000 to Php4999, …
bracket
type of work
age vendor,
35, farmer,
19, 27, 43, …fisherman, teacher, …
educational primary level, secondary level, tertiary level,
weight
attainment deep
64 kg,well, bottled
73 kg,
doctorate, water,
… 167 lb, 155 NAWASA,
kg, … river
water source
water, …
numberfamily class
of children A, B, 3,
1, 2, C, 4,
D,… …
kind of illness fever, cough, sore eyes, colds, …
monthly pension
awareness to Php6000, Php9500, Php4500, Php5200, …
yes, no
treatments
Sex male, female
types of CR sealed, septic, latrine, others, …
Status married, single, widow, …
Variable Classification
labels for VARIABLES are entered here
The Interface
Nominal variable
examples possible raw data codes SPSS entries
awareness to treatments “0” = no, “1” = yes 0, 1
sex of family head “0” = female, “1” = male 0, 1
type of CR “0” = none, “1” septic, “2” = sealed, …0, 1, 2, …
current illness “0” = fever, “1” = dengue, “2” = TB, … 0, 1,
2, …
water source “0” = deep well, “1” = NAWASA, … 0, 1, 2, …
Ordinal variable
examples possible raw data codes SPSS entries
birth order “1” = first, “2” = second, “3” =third, … 1, 2, 3, …
monthly salary bracket “1” = < P 1000, “2” = P 1000 - P 4999, … 1,
2, 3, …
educational attainment “1” = elementary, “2” = high school, … 1,
2, 3, …
class of family “1” = A, “2” = B, “3” = C, “4” = DEF 1, 2, 3, 4
high school year “1” = freshman, “2” = sophomore, … 1, 2,
3, 4
Entering data
For the first set of data (Househld I.D.
Number), data codes are nominal. [In
SPSS, specify the variable as “string.”]
HOUSEHOL
D I.D.
NUMBER
001
002
003
004
005
006
007
008
009
010
Entering data
For the second set of data
(baranggay), values are also nominal.
Say that Tibanga is coded as “1” in
SPSS,
HOUSEHOLthe data are entered as:
D I.D.
BARANGGA
Y
NUMBER
001 Tibanga
002 Tibanga
003 Tibanga
004 Tibanga
005 Tibanga
006 Tibanga
007 Tibanga
008 Tibanga
009 Tibanga
010 Tibanga
Entering data
Sex is also nominal. If the user codes
“1” for male and “2” for female:
HOUSEHOL SEX OF
D I.D. FAMILY
NUMBER HEAD
001 male
002 male
003 female
004 male
005 female
006 male
007 male
008 male
009 male
010 male
Entering data
Data values for illnesses are also
nominal. We designate “1” for cough;
“2” for fever; “3”, colds; “4”,
COMMON
toothache,
HOUSEHOL and; “5” for stomachache.
ILLNESS
D I.D.
EXPERIEN
NUMBER
CED
001 cough
002 fever
003 colds
004 colds
005 cough
006 toothache
stomachac
007
he
008 cough
009 fever
010 fever
Entering data
The family’s salary can be expressed
as a numerical value. The data can be
entered as is:
HOUSEHOL SALARY OF
D I.D. HOUSEHO
NUMBER LD
001 Php 1500
002 Php 800
003 Php 2000
004 Php 2500
005 Php 3000
006 Php 1100
007 Php 5000
008 Php 8000
009 Php 3000
010 Php 2000
Entering data
On the other hand, if the user has set salary
brackets, the data becomes ordinal. Say the
brackets are set up as “1” = below Php 1000; “2” =
Php 1000 to Php 2999; “3” = Php 3000 to Php 6999;
HOUSEHOL
“4” = Php 7000SALARY
andOFabove, the data are entered in
D I.D. HOUSEHO
SPSS as
NUMBER follows: LD
001 Php 1500
002 Php 800
003 Php 2000
004 Php 2500
005 Php 3000
006 Php 1100
007 Php 5000
008 Php 8000
009 Php 3000
010 Php 2000
Entering data
Finally, for the family head’s work
(which are nominal), let “1” =
fisherman; “2” = carpenter; “3” =
HOUSEHOL WORK OF
laborer;
D I.D. “4” = vendor; “5” = others
FAMILY
NUMBER HEAD
001 fisherman
002 carpenter
003 fisherman
004 fisherman
005 laborer
006 vendor
007 laborer
008 others
009 laborer
010 vendor
Entering data
Final Output
Frequency Distribution of
Manufacturer
Revisit: Skewness and Kurtosis
• Two sets of data may have the same
mean and standard deviation but
they vary on shape.
• +/-2 is acceptable for both skewness
and kurtosis
• Rule of Thumb:
less than -1 or greater than 1, the
distribution is highly skewed
between -1 and -1/2 or between ½
and 1, the distribution is moderately
skewed
between -1/2 and ½, the distribution