*******____WATER CONSUMPTON DETERMINANTS ARUBA AND THE 
NETHERLANDS____*******.
*Update: 4 december 2022.

****NORMALITY OF DISTRIBUTION CHECK****. 
FREQUENCIES VARIABLES=WAT_CONS SSB_CONS BI_WAT AT_WAT SE_WAT 
DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
   /STATISTICS=STDDEV VARIANCE RANGE MINIMUM MAXIMUM SEMEAN MEAN 
MEDIAN SKEWNESS SESKEW KURTOSIS 
    SEKURT
  /HISTOGRAM NORMAL
  /ORDER=ANALYSIS.

COMPUTE SQRTSSB_CONS=SQRT(SSB_CONS).
EXECUTE.

FREQUENCIES VARIABLES= SQRTSSB_CONS 
  /STATISTICS=STDDEV VARIANCE RANGE MINIMUM MAXIMUM SEMEAN MEAN MEDIAN 
SKEWNESS SESKEW KURTOSIS 
    SEKURT
  /HISTOGRAM NORMAL
  /ORDER=ANALYSIS.

****DESCRIPTIVE STATISTICS****. 
CROSSTABS
  /TABLES=Country BY School_type
  /FORMAT=AVALUE TABLES
  /CELLS=COUNT TOTAL 
  /COUNT ROUND CELL.

*Descriptives for entire sample.
DESCRIPTIVES VARIABLES=School_type Sex Age Country TL WAT_CONS SSB_CONS 
BI_WAT AT_WAT
SE_WAT DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /STATISTICS=MEAN STDDEV MIN MAX.

FREQUENCIES VARIABLES=School_type Sex Age Country TL WAT_CONS SSB_CONS 
BI_WAT AT_WAT
SE_WAT DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /STATISTICS=STDDEV MINIMUM MAXIMUM MEAN
  /ORDER=ANALYSIS.

*Descriptives for each separate country.
SPLIT FILE SEPARATE BY Country.

DESCRIPTIVES VARIABLES=School_type Sex Age TL SSB_CONS WAT_CONS  BI_WAT 
AT_WAT
SE_WAT DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
   /STATISTICS=MEAN STDDEV MIN MAX.

FREQUENCIES VARIABLES=School_type Sex Age TL SSB_CONS WAT_CONS  BI_WAT 
AT_WAT
SE_WAT DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /STATISTICS=STDDEV MINIMUM MAXIMUM MEAN
  /ORDER=ANALYSIS.

SPLIT FILE OFF.

*INDEPENDENT SAMPLES T-TEST*.
T-TEST GROUPS=Country(0 1)
  /MISSING=ANALYSIS
  /VARIABLES=TL SSB_CONS WAT_CONS BI_WAT AT_WAT SE_WAT DN_P_WAT 
DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /CRITERIA=CI(.95).

T-TEST GROUPS=Country(0 1)
  /MISSING=ANALYSIS
  /VARIABLES=Age Sex School_type
  /CRITERIA=CI(.95).

T-TEST GROUPS=Country(0 1)
  /MISSING=ANALYSIS
  /VARIABLES=Age Sex
  /CRITERIA=CI(.95).

****CORRELATIONS****. 
SPLIT FILE SEPARATE BY Country.

CORRELATIONS
  /VARIABLES=TL Age SSB_CONS WAT_CONS BI_WAT AT_WAT SE_WAT DN_P_WAT 
DN_F_WAT 
    IN_P_WAT IN_F_WAT IM_WAT
  /PRINT=TWOTAIL NOSIG
  /MISSING=PAIRWISE. 

SPLIT FILE OFF.

****PREPARATION FOR HIERARCHICAL MULTIPLE REGRESSION ANALYSES****.
*Mean-center the determinants. 
*Then compute interaction variables between country of residence and the mean-
centered behavioral determinants. 
DESCRIPTIVES VARIABLES=BI_WAT AT_WAT SE_WAT DN_P_WAT DN_F_WAT IN_P_WAT 
IN_F_WAT IM_WAT
  /STATISTICS=MEAN STDDEV MIN MAX.

compute cBI_WAT = BI_WAT - 4.59.
compute cAT_WAT=AT_WAT - 3.33.
compute cSE_WAT = SE_WAT - 4.94.
compute cDN_P_WAT = DN_P_WAT - 4.82.
compute cDN_F_WAT = DN_F_WAT - 4.01.
compute cIN_P_WAT = IN_P_WAT - 4.98.
compute cIN_F_WAT = IN_F_WAT - 3.61.
compute cIM_WAT = IM_WAT - 4.52.
execute.

compute ctryXBI_WAT = cBI_WAT * Country.
compute ctryXAT_WAT = cAT_WAT * Country.
compute ctryXSE_WAT = cSE_WAT * Country.
compute ctryXDN_P_WAT = cDN_P_WAT * Country.
compute ctryXDN_F_WAT = cDN_F_WAT * Country.
compute ctryXIN_P_WAT = cIN_P_WAT * Country.
compute ctryXIN_F_WAT = cIN_F_WAT * Country.
compute ctryXIM_WAT = cIM_WAT * Country.
execute.

****HIERARCHICAL MULTIPLE REGRESSION FOR ARUBA****. 
USE ALL.
COMPUTE filter_$=(Country = 0).
VARIABLE LABELS filter_$ 'Country = 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N  
 /MISSING LISTWISE
  /STATISTICS COEFF OUTS CI(95) R ANOVA COLLIN TOL CHANGE ZPP
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN 
  /DEPENDENT WAT_CONS
  /METHOD=ENTER TL Sex Age SSB_CONS BI_WAT AT_WAT SE_WAT DN_P_WAT 
DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /PARTIALPLOT ALL
  /SCATTERPLOT=(*ZRESID ,*ZPRED)
  /RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
  /CASEWISE PLOT(ZRESID) OUTLIERS(2).

FILTER OFF.
USE ALL.
EXECUTE.

****HIERARCHICAL MULTIPLE REGRESSION FOR TOTAL SAMPLE***. 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N  
 /MISSING LISTWISE
  /STATISTICS COEFF OUTS CI(95) R ANOVA COLLIN TOL CHANGE ZPP
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN 
  /DEPENDENT WAT_CONS
  /METHOD=ENTER TL Sex Age SSB_CONS Country BI_WAT AT_WAT SE_WAT DN_P_WAT 
DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT ctryXBI_WAT ctryXAT_WAT ctryXSE_WAT 
ctryXDN_P_WAT ctryXDN_F_WAT 
    ctryXIN_P_WAT ctryXIN_F_WAT ctryXIM_WAT 
  /PARTIALPLOT ALL
  /SCATTERPLOT=(*ZRESID ,*ZPRED)
  /RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
  /CASEWISE PLOT(ZRESID) OUTLIERS(2).

****MODERATION ANALYSES****. 
*The regression analyses showed two significant interaction terms: 
ctryXIM_WAT (between country and intrinsic motivation) and 
ctryXDN_F_WAT (country and descriptive norm).
*These were further interpreted with Hayes' SPSS PROCESS macro version 4.0.

**INTERACTION BETWEEN COUNTRY AND INTRINSIC MOTIVATION**
"Point and Click" settings used for the significant interaction term ctryXIM_WAT
Y: WAT_CONS
X: IM_WAT 
W: Country 
Covariates: TL, Sex, Age, SSB consumption
Model Number: 1
Confidence level 95
Bootstrap samples: 1000
OPTIONS click:
Show covariance matrix of regression coefficients
Generate code for visualizing interactions
Pairwise contrasts of indirect effects
Test for X by M interaction(s)
Heteroscedasticity-consistent inference: none
Decimal places in output: 4
Mean center for construction of products: Only continuous variables that define 
products
Moderation and conditioning. Probe interactions if p < .05
Conditioning values: -1SD, Mean, +1SD
MULTICATIGORICAL: none 

*The "data for visualizing conditional effect of X on Y" derived from the moderation 
analysis was used
to plot Figure 1 in Excel. The data list below was used in Excel except row 2 and row 5, 
so that only the low and high levels were plotted. 
*The figures/graphs were plotted with the syntax commando and its lay-out was 
improved using Excel.

Data for visualizing conditional effect of X on Y 
Paste text below into a SPSS syntax window and execute to produce plot. 

 DATA LIST FREE/ 
   IM_WAT     Country    WAT_CONS   . 
BEGIN DATA. 
    -1.2745      .0000     3.4081 
      .0000      .0000     4.2275 
     1.2745      .0000     5.0470 
    -1.2745     1.0000     2.9661 
      .0000     1.0000     3.3204 
     1.2745     1.0000     3.6747 
END DATA. 
GRAPH/SCATTERPLOT= 
 IM_WAT   WITH     WAT_CONS BY       Country  .

**INTERACTION BETWEEN COUNTRY AND PERCEIVED DESCRIPTIVE NORM 
FRIENDS**
"Point and Click" settings used for the significant interaction term ctryDN_F_WAT
Y: WAT_CONS
X: DN_F_WAT  
W: Country
Covariates: TL, Sex, Age, SSB consumption
*The other point and click settings are the same as for the previous moderation 
analysis.
*Figure 2 was plotted in the same way as the previous moderation analysis with the 
following data list. 

DATA LIST FREE/ 
   DN_F_WAT   Country    WAT_CONS   . 
BEGIN DATA. 
    -1.2860      .0000     3.9013 
      .0000      .0000     4.2489 
     1.2860      .0000     4.5965 
    -1.2860     1.0000     3.0042 
      .0000     1.0000     3.1551 
     1.2860     1.0000     3.3060 
END DATA. 
GRAPH/SCATTERPLOT= 
 DN_F_WAT WITH     WAT_CONS BY       Country  .

****HIERARCHICAL MULTIPLE REGRESSION FOR ARUBA WITH TRANSFORMED SSB 
VARIABLE ****. 
USE ALL.
COMPUTE filter_$=(Country = 0).
VARIABLE LABELS filter_$ 'Country = 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N  
 /MISSING LISTWISE
  /STATISTICS COEFF OUTS CI(95) R ANOVA COLLIN TOL CHANGE ZPP
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN 
  /DEPENDENT WAT_CONS
  /METHOD=ENTER TL Sex Age SQRTSSB_CONS BI_WAT AT_WAT SE_WAT DN_P_WAT 
DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT
  /PARTIALPLOT ALL
  /SCATTERPLOT=(*ZRESID ,*ZPRED)
  /RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
  /CASEWISE PLOT(ZRESID) OUTLIERS(2).

FILTER OFF.
USE ALL.
EXECUTE.

****HIERARCHICAL MULTIPLE REGRESSION FOR TOTAL SAMPLE WITH TRANSFORMED 
SSB VARIABLE***. 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N  
 /MISSING LISTWISE
  /STATISTICS COEFF OUTS CI(95) R ANOVA COLLIN TOL CHANGE ZPP
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN 
  /DEPENDENT WAT_CONS
  /METHOD=ENTER TL Sex Age SQRTSSB_CONS Country BI_WAT AT_WAT SE_WAT 
DN_P_WAT DN_F_WAT IN_P_WAT IN_F_WAT IM_WAT ctryXBI_WAT ctryXAT_WAT 
ctryXSE_WAT ctryXDN_P_WAT ctryXDN_F_WAT 
    ctryXIN_P_WAT ctryXIN_F_WAT ctryXIM_WAT 
  /PARTIALPLOT ALL
  /SCATTERPLOT=(*ZRESID ,*ZPRED)
  /RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
  /CASEWISE PLOT(ZRESID) OUTLIERS(2).

*_______END_______*