USING SPSS SYNTAX CODES

SPSS SYNTAX CODES –Keyed to de Vaus Chapters

*Chapter 3*.
*Uses CRIC 2002 Charter Data*.

*Basic Recodes*.

recode q13 (1=4) (2=3) (3=2) (4=1) into Chrtgood.
value labels Chrtgood 1 'vbad' 2 'bad' 3 'good' 4 'vgood'.

recode q37 (1=2) (2=1) into Chrtunit.
value labels Chrtunit 1 'divide' 2 'unite'.

*Using deviation coding*.
recode q16a (1= -1)(2=0)(3=1) into ChrtMin.
recode q16b (1= -1)(2=0)(3=1)into ChrtCan.
recode q16c (1= -1)(2=0)(3=1)into ChrtWom.
recode q16d (1= -1)(2=0)(3=1)into ChrtCrim.
value labels ChrtMin ChrtCan ChrtWom ChrtCrim -1 'too far' 0 'just enough' 1 'not enough' .

*Using dichotomous coding*.
recode q16a (1,2=1) (3=2) into ChrtMin2.
recode q16b (1,2=1) (3=2) into ChrtCan2.
recode q16c (1,2=1) (3=2) into ChrtWom2.
recode q16d (1,2=1) (3=2) into ChrtCrm2.
value labels ChrtMin2 ChrtCan2 ChrtWom2 ChrtCrm2 1 'too far or not enough' 2 'just enough'.

*Cohen's Kappa - "not of much use" per deVaus p19*.
CROSSTABS TABLES=ChrtMin BY ChrtWom
   /FORMAT=NOTABLES
   /STATISTIC=KAPPA
   /COUNT ROUND CELL .

*Inter-item Correlations -Using taub in lieu of Pearson's r*.
NONPAR CORR VARIABLES=ChrtMin ChrtCan ChrtWom ChrtCrim
   /PRINT=KENDALL TWOTAIL SIG
   /MISSING=PAIRWISE .

*Reliability Analyses*.

*Using deviation coding*.
RELIABILITY
   /VARIABLES= Chrtgood Chrtunit ChrtMin ChrtCan ChrtWom
   /SCALE(ALPHA)=ALL/MODEL=ALPHA
   /STATISTICS=DESCRIPTIVE
   /SUMMARY=TOTAL .

*Using dichotomous coding*.
RELIABILITY
   /VARIABLES=Chrtgood Chrtunit ChrtMin2 ChrtCan2 ChrtWom2 ChrtCrm2
   /SCALE(ALPHA)=ALL
   /MODEL=ALPHA
   /STATISTICS=DESCRIPTIVE
   /SUMMARY=TOTAL .

*Split-Halves Method*.
RELIABILITY
   /VARIABLES=Chrtgood Chrtunit ChrtMin ChrtCan ChrtWom ChrtCrim
   /SCALE(SPLIT)=ALL/MODEL=SPLIT.

*Cronbach's Alpha*.
RELIABILITY
   /VARIABLES=Chrtgood Chrtunit ChrtMin ChrtCan ChrtWom ChrtCrim
   /SCALE(ALPHA)=ALL/MODEL=ALPHA
   /SUMMARY=TOTAL .

* de Vaus Chapter 4*.
*Correlation Matrix-- again using taub in lieu of r*.
NONPAR CORR
   /VARIABLES=Chrtgood Chrtunit ChrtMin ChrtCan ChrtWom ChrtCrim
   /PRINT=KENDALL TWOTAIL SIG
   /MISSING=PAIRWISE .

deVaus Chapters 5, 6, 8, 9, 28, 30, 32 & 36.

The data source for the following examples is the IRPP Court and the Charter Survey. The responses to two questions in the survey are used as examples for this week:

Q44. What was the total income BEFORE taxes of ALL MEMBERS of your household for 1998. I do not need the exact amount, just a broad category.

[PAUSE] Was the total household income from all sources…

<1> less than $10,000

<2> $10,000 to less than $20,000

<3> $20,000 to less than $30,000

<4> $30,000 to less than $40,000

<5> $40,000 to less than $50,000

<6> $50,000 to less than $60,000

<7> $60,000 to less than $70,000

<8> $70,000 to less than $80,000

<9> $80,000 to less than $90,000

<10>$90,000 to less than $100,000

<11> or over $100,000

Q18. From what you have heard or read, would you say that you are very satisfied, somewhat satisfied, not very satisfied or not satisfied at all with the way the Supreme Court has been working?

<1> Very satisfied

<2> Somewhat satisfied

<4> Not very satisfied

<5> Not satisfied at all

<8> Don’t know

<9> Refused

Chapter 5 Exercises

First, exclude all cases from q44 where respondents said they "don't know" or "refused" (see Chapter 9). The SPSS syntax is as follows:

MISSING VALUES q44 (98,99).

SPSS Syntax for Obtaining a Frequency Table and Cutting Points

FREQUENCIES

VARIABLES=q44

/NTILES= 3 .

SPSS Syntax for Recoding a Variable

RECODE

q44

(1 thru 3=1) (4 thru 5=2) (6 thru 11=3) INTO income .

EXECUTE .

Chapter 6 Exercises

Again, exclude all cases from q44 where respondents said they "don't know" or "refused" (see Chapter 9). The SPSS syntax is as follows:

missing values q44 (98,99).

SPSS Syntax for Converting Variables to the Interval Level

RECODE

q44

(1=5000)(2=15000)(3=25000)(4=35000)(5=45000)(6=55000)

(7=65000)(8=75000)(9=85000)(10=95000)(11=105000)

INTO income .

EXECUTE .

Chapter 8 Exercises

SPSS Syntax for Reordering and Reversing Categories

RECODE

q18

(1=5) (2=4) (4=2) (5=1) INTO q18rev.

EXECUTE .

Chapter 9 Exercises

SPSS Syntax for Dropping Cases

MISSING VALUES q18 (8,9).

SPSS Syntax for Imputation

RMV

/q18_1=SMEAN(q18).

Chapter 28 Exercises

SPSS Syntax for Producing Frequency Tables

FREQUENCIES

VARIABLES=q18.

Chapter 30

SPSS Syntax for Descriptive Univariate Summary Statistics

MISSING VALUES q18 (8,9).

FREQUENCIES

VARIABLES=q18 /FORMAT=NOTABLE

/STATISTICS=STDDEV VARIANCE RANGE MEAN MEDIAN MODE SKEWNESS SESKEW KURTOSIS

SEKURT.

Chapter 32 & 36

SPSS Syntax for Crosstabs with Nominal & Ordinal Data from CRIC 2002

Crosstabs with Dichotomous Variables (refugees and civil libs)- using phi.

missing values q25 q35 (99).
crosstabs tables = q25 q35 by lang gender
/cells = column count
/statistics = phi.

Additional Dichotomies on porn and hate law from CRIC 2002

missing values q23_1 q23_2 (88, 99).
crosstabs tables = q23_1 q23_2 by lang gender
      /cells = column count
      /statistics = phi.

*Multi-nomial independent variables (region & occup) - uses V*.

missing values q25 q35 (99).
crosstabs tables = q25 q35 by region q43
    /cells = column count
    /statistics = phi.

Ordinal crosstabs with CRIC 2002 data using taub & tauc.

missing values q13 q42 (99).
recode q13 (1=4) (2=3) (3=2) (4=1) into chart4.
value labels chart4 1 ' vbad' 2 'bad' 3 'good' 4 'vgood'.
recode q42 (1,2,3=1) (4,5,6 = 2) (7,8=3) into educ.
value labels educ 1 'HS or less' 2 'Some PS' 3 'Univ Grad'.

crosstabs tables = chart4 by educ
/cells = column count
/statistics = btau ctau gamma.

Chapter 39

Syntax for Testing Statistical Significance

Using the 2000 Canadian Election Study

T-Test

missing values PESC3 (998, 999).

fre var=PESC3 PESLANG/.

t-test groups = PESLANG (1, 2)

/variables = PESC3

/criteria = CIN (.99).

One-Way ANOVA

missing values CPSPC (996, 998, 999).

fre var=CPSPC PROVINCE/.

recode PROVINCE (10, 12, 13=1) (24=2) (35=3) (46, 47, 48=4)

(59, 11=5) into region.

value labels region 1'Atlantic' 2'Quebéc' 3'Ontario'

4'Prairies' 5'West'.

fre var= region.

Oneway CPSPC by region (1, 5)

/ranges = scheffe

/statistics = all.

Chi-Square Test

missing values CPSA2D (7, 8, 9).

fre var=CPSA2D CPSM10B/.

recode CPSA2D (1=3) (2=2) (3=1) into famval.

value labels famval 1'not important' 2'smw important'

3'very important'.

recode CPSM10B (1=4) (3=3) (5=2) (7=1) into religion.

value labels religion 1'not at all' 2'not important'

3'important' 4'very important'.

fre var=famval religion/.

crosstab tables = famval by religion

/cells=column count

/statistics = CHISQ ctau.