Appendix – hierarchical linear regression output from Kiwis Count pilot data - Te Kawa Mataaho Public Service Commission

Regression results using Trust_Brand as the criterion

Predictor	b	b_95%_CI	sr2	sr2_95%_CI
(Intercept)	3.49**	[3.22, 3.75]
factor_gender2	-0.09*	[-0.18, -0.00]	.00	[-.00, .01]
factor_gender3	0.24	[-1.62, 2.10]	.00	[-.00, .00]
factor_Age4	-0.14	[-0.32, 0.03]	.00	[-.00, .00]
factor_Age5	-0.15	[-0.33, 0.04]	.00	[-.00, .00]
factor_Age6	-0.02	[-0.20, 0.16]	.00	[-.00, .00]
factor_Age7	0.12	[-0.09, 0.33]	.00	[-.00, .00]
Inc_agg$50,001-$100,000	0.12*	[0.01, 0.24]	.00	[-.00, .01]
Inc_agg>$100,000	0.09	[-0.03, 0.21]	.00	[-.00, .00]
factor_Disability2	0.13*	[0.01, 0.25]	.00	[-.00, .01]
factor_Education2	0.15	[-0.04, 0.33]	.00	[-.00, .00]
factor_Education3	0.15	[-0.03, 0.34]	.00	[-.00, .00]
factor_Education4	0.14	[-0.04, 0.33]	.00	[-.00, .00]
factor_Education5	1.28	[-0.58, 3.14]	.00	[-.00, .00]
QC_G1	-0.06	[-0.20, 0.09]	.00	[-.00, .00]
QC_G2	-0.04	[-0.17, 0.09]	.00	[-.00, .00]
QC_G3	0.07	[-0.10, 0.24]	.00	[-.00, .00]
QC_G4	0.17	[-0.01, 0.35]	.00	[-.00, .01]
QC_G5	-0.09	[-0.38, 0.19]	.00	[-.00, .00]
QC_G6	0.04	[-0.13, 0.20]	.00	[-.00, .00]
				R2 = .027**
				95% CI[.00,.03]
(Intercept)	0.88**	[0.63, 1.14]
factor_gender2	-0.06	[-0.13, 0.01]	.00	[-.00, .00]
factor_gender3	-0.11	[-1.55, 1.33]	.00	[-.00, .00]
factor_Age2	0.05	[-0.07, 0.17]	.00	[-.00, .00]
factor_Age3	-0.03	[-0.15, 0.09]	.00	[-.00, .00]
factor_Age4	-0.05	[-0.19, 0.08]	.00	[-.00, .00]
factor_Age5	0.01	[-0.14, 0.15]	.00	[-.00, .00]
factor_Age6	0.08	[-0.06, 0.22]	.00	[-.00, .00]
factor_Age7	0.12	[-0.04, 0.28]	.00	[-.00, .00]
Inc_agg$50,001-$100,000	0.08	[-0.01, 0.17]	.00	[-.00, .00]
Inc_agg>$100,000	0.06	[-0.04, 0.15]	.00	[-.00, .00]
factor_Disability2	0.13**	[0.04, 0.23]	.00	[-.00, .01]
factor_Education2	0.04	[-0.10, 0.19]	.00	[-.00, .00]
factor_Education3	0.10	[-0.05, 0.24]	.00	[-.00, .00]
factor_Education4	0.08	[-0.07, 0.22]	.00	[-.00, .00]
factor_Education5	1.15	[-0.28, 2.59]	.00	[-.00, .00]
QC_G1	-0.03	[-0.14, 0.08]	.00	[-.00, .00]
QC_G2	-0.03	[-0.14, 0.07]	.00	[-.00, .00]
QC_G3	-0.01	[-0.14, 0.12]	.00	[-.00, .00]
QC_G4	0.07	[-0.07, 0.21]	.00	[-.00, .00]
QC_G5	0.00	[-0.22, 0.23]	.00	[-.00, .00]
QC_G6	0.00	[-0.12, 0.13]	.00	[-.00, .00]
row_mean 0.74**	[0.70, 0.78]	[0.70, 0.78]	.39	[.36, .43]
				R2 = .420 Delta R2 = .393
				95% CI[.38,.44] 95% CI[.36, .43]

Note. A significant b-weight indicates the semi-partial correlation is also significant.

b represents unstandardized regression weights.

sr2 represents the semi-partial correlation squared.

Square brackets are used to enclose the lower and upper limits of a confidence interval.

* indicates p < .05. ** indicates p < .01.

> anova(KC1P) Analysis of Variance Table

Response: Trust_Brand
	Df	Sum Sq	Mean Sq	F value	Pr(>F)
factor_gender	2	5.89	2.9456	3.3510	0.035251	*
factor_Age	6	8.08	1.3472	1.5326	0.163532
Inc_agg	3	11.26	3.7531	4.2697	0.005161	**
factor_Disability	1	6.43	6.4271	7.3117	0.006910	**
factor_Education	4	2.94	0.7338	0.8348	0.502937
QC_G1	1	2.73	2.7272	3.1025	0.078327	.
QC_G2	1	2.09	2.0898	2.3774	0.123267
QC_G3	1	0.08	0.0809	0.0920	0.761688
QC_G4	1	4.23	4.2331	4.8157	0.028319	*
QC_G5	1	0.29	0.2872	0.3267	0.567689
QC_G6	1	0.16	0.1620	0.1843	0.667735
Residuals	1943 -1707.93 - 0.8790
Signif. codes: 0 ‘*’ 0.001 ‘’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> anova(KC4P) Analysis of Variance Table

Response: Trust_Brand
	Df	Sum Sq	Mean Sq	F value	Pr(>F)
factor_gender	2	5.89	2.95	5.6808	0.0034679	**
factor_Age	6	8.08	1.35	2.5981	0.0164336	*
Inc_agg	3	11.26	3.75	7.2381	7.903e-05	***
factor_Disability	1	6.43	6.43	12.3951	0.0004404	***
factor_Education	4	2.94	0.73	1.4152	0.2264128
QC_G1	1	2.73	2.73	5.2595	0.0219330	*
QC_G2	1	2.09	2.09	4.0302	0.0448305	*
QC_G3	1	0.08	0.08	0.1560	0.6929521
QC_G4	1	4.23	4.23	8.1638	0.0043191	**
QC_G5	1	0.29	0.29	0.5538	0.4568625
QC_G6	1	0.16	0.16	0.3125	0.5762367
row_mean	1	700.96	700.96	1351.8422	< 2.2e-16	***
Residuals	1942 - 1006.97 - 0.52
Signif. codes:	0 ‘*’ 0.001 ‘’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Appendix – hierarchical linear regression output from Kiwis Count pilot data

Measuring the drivers of public trust: An exploratory study

Appendix - pilot data

Regression results using Trust_Brand as the criterion

> anova(KC1P) Analysis of Variance Table

> anova(KC4P) Analysis of Variance Table