Analysis Report
Study 1
ModelSTD11:
A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD11, RMSEA = 0.049, CFI > 0.946, and TLI = 0.939, which are very close to good model fit values. We can conclude that ModelSTD11 is a good model fit.
Social norm value is the only predictor that affects the Collectivism, Individualism, and environmental concern in a positive sense. Moreover, PCB is positively affected by INTGP.

ModelSTD11:
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
Collectivism ~
SN 0.353 0.085 4.171 0 0.347 0.347
ATTSmart 0.157 0.097 1.621 0.105 0.198 0.198
AttNoPlastic -0.139 0.124 -1.126 0.26 -0.132 -0.132
Individulaism ~
SN 0.172 0.077 2.234 0.025 0.164 0.164
ATTSmart 0.075 0.091 0.823 0.411 0.092 0.092
AttNoPlastic 0.12 0.117 1.032 0.302 0.111 0.111
EnvirConcern ~
SN 0.451 0.076 5.895 0 0.439 0.439
ATTSmart 0.03 0.084 0.357 0.721 0.037 0.037
AttNoPlastic 0.166 0.107 1.545 0.122 0.156 0.156
PBC ~
INTGP 0.703 0.053 13.242 0 0.692 0.692

ModelSTD12
A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD12, RMSEA = 0.044, CFI > 0.983, and TLI = 0.977, which are very close to good model fit values. We can conclude that ModelSTD12 is a good model fit.

ImpBEnvFr is positively affected by Collectivism and Individualism. On the contrary, BarrBEnvFr is negatively affected by Collectivism and Indvidualism.

ModelSTD12
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
ImpBEnvFr ~
Collectivism 0.608 0.083 7.344 0 0.475 0.475
Individulaism 0.289 0.071 4.055 0 0.234 0.234
BarrBEnvFr ~
Collectivism -0.253 0.075 -3.345 0.001 -0.222 -0.222
Individulaism -0.445 0.073 -6.129 0 -0.405 -0.405

ModelSTD13
A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD13, RMSEA = 0.044, CFI > 0.969, and TLI = 0.963, which are very close to good model fit values. We can conclude that ModelSTD13 is a good model fit.
ImpBEnvFr is affecting SN, PBC, AttNoPlastic, and ATTSmart positively by BarrBEnvFr is affecting them negatively.

ModelSTD13
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
SN ~
ImpBEnvFr 0.389 0.054 7.194 0 0.513 0.513
BarrBEnvFr -0.231 0.06 -3.855 0 -0.266 -0.266
PBC ~
ImpBEnvFr 0.2 0.056 3.571 0 0.221 0.221
BarrBEnvFr -0.457 0.07 -6.501 0 -0.44 -0.44
AttNoPlastic ~
ImpBEnvFr 0.176 0.054 3.272 0.001 0.235 0.235
BarrBEnvFr -0.206 0.064 -3.22 0.001 -0.241 -0.241
ATTSmart ~
ImpBEnvFr 0.287 0.064 4.498 0 0.291 0.291
BarrBEnvFr -0.331 0.077 -4.325 0 -0.293 -0.293

ModelSTD14:
A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD14, RMSEA = 0.054, CFI > 0.953, and TLI = 0.945, which are very close to good model fit values. We can conclude that ModelSTD14 is a good model fit.
ATTSmart, AttNoPlastic, and EnvirConcern are positively affecting INTGP, where AttNoPlastic is the strongest one.
ATTSmart and EnvirConcern are positively related to SN and AttNoPlastic has no relation with SN.
PBC is also positively affected by all the three with EnvirConcern with the strongest relation.

ModelSTD14
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
INTGP ~
ATTSmart 0.189 0.073 2.584 0.01 0.209 0.209
AttNoPlastic 0.838 0.108 7.727 0 0.698 0.698
EnvirConcern 0.098 0.045 2.187 0.029 0.088 0.088
SN ~
ATTSmart 0.294 0.083 3.533 0 0.375 0.375
AttNoPlastic 0.052 0.111 0.472 0.637 0.05 0.05
EnvirConcern 0.387 0.058 6.726 0 0.403 0.403
PBC ~
ATTSmart 0.216 0.082 2.615 0.009 0.235 0.235
AttNoPlastic 0.315 0.113 2.794 0.005 0.258 0.258
EnvirConcern 0.4 0.056 7.169 0 0.356 0.356

Study 2
ModelSTD21 <- lm(IntUsePCUP ~ FVQ + FVP + SV + EmV + CV + EpV) Table: STD21 (Linear Regression Model) Coefficients: Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.43637 1.3305 1.831 0.067789 .
FVQ 0.36415 0.06221 5.853 9.77e-09 ***
FVP 0.14224 0.07076 2.010 0.045044 *
SV 0.17980 0.05885 3.055 0.002392 **
EmV 0.32916 0.07504 4.386 1.46e-05 ***
CV 0.08471 0.03394 2.496 0.012941 *
EpV 0.18991 0.04948 3.838 0.000143 ***

IntUsePCUP is affected by all the independent variables. FVQ is affecting the most and CV is affecting the least. Adjusted R-square indicates that 34% of the variation in IntUsePCUP is explained by these independent variables. Increase in all the values, quality, price, social, Emotional, conditional, and epistemic will enhance the use of public cup.

Model: STD22 <- lm(DI ~ FVQ + FVP + SV + EmV + CV + EpV) INTGP <- INTGP1 + INTGP2 + INTGP3 DI <- IntUsePCUP - INTGP Table: STD22 (Linear Regression Model) Coefficients Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.54674 1.40521 -1.101 0.27166
FVQ 0.18044 0.06571 2.746 0.00629 **
FVP 0.05409 0.07473 0.724 0.46957
SV 0.12172 0.06215 1.958 0.05085 .
EmV 0.0912 0.07925 1.151 0.25049
CV 0.02069 0.03584 0.577 0.56413
EpV 0.22517 0.05226 4.309 2.05e-05 ***

The variable DI is calculated by taking the difference between IntUsePCUP and INTGP. DI is affected by EpV and FVQ positively. The other four are found insignificant. In case of DI, only two values, quality and Epistemic with high score will produce high DI score. The other four will not change the DI significantly. Adjusted R-square indicating 10% explanation.
ModelSTD23:
Table: STD23: (Structural Equation Model)
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
IntUsePCUP ~
FVQ 0.190 0.047 4.030 0.000 0.252 0.252
FVP 0.032 0.036 0.878 0.380 0.049 0.049
SV 0.071 0.028 2.497 0.013 0.144 0.144
EmV 0.087 0.046 1.865 0.062 0.115 0.115
CV 0.029 0.022 1.340 0.180 0.060 0.060
EpV 0.120 0.031 3.900 0.000 0.188 0.188
INTGP ~
FVQ 0.239 0.073 3.262 0.001 0.228 0.228
FVP 0.066 0.059 1.108 0.268 0.073 0.073
SV 0.053 0.046 1.158 0.247 0.077 0.077
EmV 0.258 0.074 3.464 0.001 0.246 0.246
CV 0.070 0.036 1.960 0.050 0.102 0.102
EpV -0.039 0.048 -0.798 0.425 -0.044 -0.044
IntUsePCUP ~
INTGP 0.224 0.041 5.473 0.000 0.312 0.312

A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD23, RMSEA = 0.073, CFI > 0.902, and TLI = 0.885, which are very close to good model fit values. We can conclude that ModelSTD23 is a good model fit.

In the given structural equation model, Intention to use PCUP is positively affected by FVQ, EpV, and SV. EmV is more specific to INTGP with FVQ. More importantly, IntUsePCUP is affected by INTGP positively. Structural equation model has provided us better information if compared to Linear Regression models STD21 and STD22. A stronger relation can be explored through mediation analysis because it seems that INTGP is mediating the relation between values and intention to use PCUP.

ModelSTD24:
ModelSTD24: Structural Equation Model
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
IntUsePCUP ~
FVQ 0.187 0.048 3.895 0.000 0.249 0.249
FVP 0.054 0.037 1.485 0.138 0.084 0.084
SV 0.092 0.029 3.186 0.001 0.187 0.187
EmV 0.101 0.046 2.168 0.030 0.135 0.135
CV 0.052 0.022 2.374 0.018 0.108 0.108
EpV 0.095 0.031 3.096 0.002 0.150 0.150
PBC ~
FVQ 0.346 0.08 4.352 0.000 0.306 0.306
FVP -0.049 0.064 -0.771 0.441 -0.050 -0.050
SV -0.060 0.049 -1.216 0.224 -0.081 -0.081
EmV 0.268 0.080 3.364 0.001 0.238 0.238
CV -0.049 0.038 -1.275 0.202 -0.066 -0.066
EpV 0.098 0.052 1.882 0.060 0.103 0.103
IntUsePCUP ~
PBC 0.159 0.033 4.776 0.000 0.240 0.240

A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD24, RMSEA = 0.064, CFI > 0.911, and TLI = 0.899, which are very close to good model fit values. We can conclude that ModelSTD23 is a good model fit.

IntUsePCUP is positively affected by all values except FVP. IntUsePCUP is mostly affected by FVQ. PBC is affected by FVQ and EmV. PBC is also affecting IntUsePCUP in a positive manner. Interestingly price does not matter as FVP is insignificant in both the relations, while quality of PCUP is the most important factor that motivates the customer to use PCUP. EmV factor cannot be ignored as it is affecting positively both PBC and IntUsePCUP. Remember that EmV is also important for INTGP in the previous model.

ModelSTD25:
ModelSTD25: Structural Equation Model
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
IntUsePCUP ~
FVQ 0.223 0.047 4.751 0.000 0.301 0.301
FVP 0.045 0.037 1.217 0.224 0.070 0.070
SV 0.100 0.030 3.396 0.001 0.205 0.205
EmV 0.073 0.049 1.477 0.140 0.098 0.098
CV 0.048 0.022 2.172 0.030 0.099 0.099
EpV 0.094 0.031 3.045 0.002 0.149 0.149
EnvConc ~
FVQ 0.100 0.071 1.418 0.156 0.099 0.099
FVP 0.010 0.058 0.180 0.857 0.012 0.012
SV -0.118 0.045 -2.599 0.009 -0.176 -0.176
EmV 0.442 0.075 5.910 0.000 0.435 0.435
CV -0.022 0.035 -0.629 0.529 -0.033 -0.033
EpV 0.099 0.048 2.078 0.038 0.115 0.115
IntUsePCUP ~
EnvConc 0.160 0.038 4.237 0.000 0.219 0.219

A good model–data fit is indicated by RMSEA < .06, CFI > .95, and TLI > .95. In case of ModelSTD25, RMSEA = 0.069, CFI > 0.905, and TLI = 0.891, which are very close to good model fit values. We can conclude that ModelSTD23 is a good model fit.

IntUsePCUP is positively affected by FVQ, SV, EpV, and CV values. IntUsePCUP is mostly affected by FVQ. EnvConc is affected by EmV, EpV, and SV. EnvConc is mostly affected by EmV. EnvConc is also affecting IntUsePCUP in a positive manner. Interestingly price does not matter as FVP is insignificant in both the relations, while EpV is the most important factor that motivates the customer to use PCUP and improve the EnvConc. SV factor has a very different effect in this model. SV is affecting IntUsePCUP positively while it affects EnvConc negatively. Consumers, who are focusing high on social values are focusing very low on Environmental concerns (Interesting).

Public Cup Preferences
PCUPA preferences are affected by FVQ and FVP. PCUPA is important for Quality concious and Price conscious customers.
PCUPB and PCUPC preferences are affected by CV. PCUPB and PCUPC is important for those customers who focus on conditional values.
PCUPD preferences are negatively affected by FVQ and FVP. PCUPD is never a good choice for Quality conscious and Price conscious customers.
I have also tried to investigate the same relationships in SPSS and got the same results. An SPSS output file is provided for reference.

M21: lm(FVQ ~ p.pcupA)
Coefficients :
Estimate Std. Error t value Pr(>|t|)
(Intercept) 15.04546 0.26437 56.911 < 2e-16 *** p.pcupA 0.5478 0.09268 5.911 7.03e-09 *** --- M22: lm(FVP ~ p.pcupA) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 11.6181 0.2376 48.9 < 2e-16 *** p.pcupA 0.28819 0.08329 3.46 0.000595 *** --- M31: lm(CV ~ p.pcupB) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 14.0795 0.457 30.81 < 2e-16 *** p.pcupB 0.4913 0.1477 3.327 0.000956 *** --- M37: lm(CV ~ p.pcupC) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 16.5304 0.3536 46.751 < 2e-16 *** p.pcupC -0.5069 0.1544 -3.283 0.00111 ** --- M39: lm(FVQ ~ p.pcupD) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 17.83164 0.2395 74.453 < 2e-16 *** p.pcupD -0.55021 0.09025 -6.096 2.45e-09 *** --- M40: lm(FVP ~ p.pcupD) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 12.99006 0.2164 60.028 < 2e-16 *** p.pcupD -0.25076 0.08155 -3.075 0.00224 ** --- Overall Intention to use Public Cup (M45 to M48) IntUsePCUP is best with PCUPB. PCUPA is also popular among the customers but slightly below PCUPB. The customers are very much reluctant to use PCUPC and PCUPD with PCUPC preference at lowest use. M45: lm(IntUsePCUP ~ p.pcupA) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 19.1373 0.3636 52.636 < 2e-16 *** p.pcupA 0.4785 0.1275 3.754 0.000199 *** --- M46: lm(IntUsePCUP ~ p.pcupB) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 18.6109 0.3716 50.086 < 2e-16 *** p.pcupB 0.6207 0.1201 5.169 3.64e-07 *** --- M47: lm(IntUsePCUP ~ p.pcupC) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 21.7279 0.2874 75.611 < 2e-16 *** p.pcupC -0.6504 0.1255 -5.183 3.38e-07 *** --- M48: M48 <- lm(IntUsePCUP ~ p.pcupD) Coefficients : Estimate Std. Error t value Pr(>|t|)
(Intercept) 21.5748 0.3298 65.41 < 2e-16 *** p.pcupD -0.4822 0.1243 -3.88 0.000121 *** --- DI = IntUsPCup-INTGP (M45 to M48) The behavior towards the use of Public Cup with reference to DI is almost the same as of IntUsePCUP.