Pretest 2 DFG Proposal | Warmth and Context

---

Connect with me on Open Science Framework | Contact me via LinkedIn

---

R analysis script for the pretest on scent selection and perception This script describes all details of the pretest. Some steps are outsourced (e.g., data labeling) to seperate scripts.

Here ist the r seassion info:

## R version 3.6.2 (2019-12-12)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 18363)
## 
## Matrix products: default
## 
## locale:
## [1] LC_COLLATE=German_Germany.1252  LC_CTYPE=German_Germany.1252   
## [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C                   
## [5] LC_TIME=German_Germany.1252    
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## loaded via a namespace (and not attached):
##  [1] compiler_3.6.2  magrittr_1.5    tools_3.6.2     htmltools_0.4.0
##  [5] yaml_2.2.0      Rcpp_1.0.1      stringi_1.4.3   rmarkdown_2.1  
##  [9] knitr_1.28      stringr_1.4.0   xfun_0.12       digest_0.6.23  
## [13] rlang_0.4.2     evaluate_0.14

Tip: include “code_folding:”show"" in header instead of “toc_float: true” & “toc_collapsed: true” and “knitr::opts_chunk$set(echo = TRUE)” to show r code

---

Notice: This part of the report only analyzes warm vs. cool scents.

1 Data preperation

1.1 Load packages thet will be used throughout the analysis

1.2 Import SPSS data from Sawtooth Lighthouse Studio

1.3 Import session protocols on temperature etc. from an Excel file

A brief overview over the imported data

Brief Overview

ScentNumber	Age	Gender	CountryNames	MDMQ1to12_r1
4	23	1	3	2
4	25	1	3	4
4	25	2	13	5
4	28	1	10	3
4	22	2	15	4
4	22	2	9	5

1.4 Rename the variables properly, clean missings, define factors and levels, and create new variables, if needed.

This is done in a seperate script (see 01_Data preparation.R)

2 Sample Description

Scent notice

	peppermint	Vanilla	p	test
n	59	65
ScentNotice = no (%)	13 (22.0)	21 (32.3)	0.280

2.1 Filter criteria

ScentSensitivity, Illness affecting sense of smell, smoking, recognition of ambient scent, effort spent on study

	peppermint	Vanilla	p	test
n	59	65
ScentSensitivity (%)			NaN
Increased	11 (22.9)	12 ( 24.5)
Normal	28 (58.3)	31 ( 63.3)
Decreased	9 (18.8)	6 ( 12.2)
Impaired	0 ( 0.0)	0 ( 0.0)
Illness = No (%)	43 (89.6)	49 (100.0)	0.063
Smoking = no (%)	41 (69.5)	54 ( 83.1)	0.116
ScentNotice = no (%)	13 (22.0)	21 ( 32.3)	0.280
Effort = LowEffort (%)	1 ( 2.1)	1 ( 2.0)	1.000

2.2 General (Net)Sample

## [1] "Sample Size"

## [1] 124

## [1] "Gender"

## 
##    Male  Female Diverse 
##   56.45   42.74    0.81

## [1] "Federal State"

## 
##    BW    BY    BE    BB    HB    HH    HE    MV    NI    NW    RP    SL    SN 
##  1.61  5.65  6.45  3.23  0.81  0.81  3.23  2.42 21.77  5.65  2.42  0.81  7.26 
##    ST    SH    TH other 
## 29.84  4.84  2.42  0.81

## [1] "Scent Sensitivity"

## 
## Increased    Normal Decreased  Impaired 
##     23.71     60.82     15.46      0.00

## [1] "Smokers"

## 
##   yes    no 
## 23.39 76.61

## [1] "Job Status"

## 
## StudyPupilApprenticeship                 Employee    SelfEmployedFreelance 
##                    94.35                     5.65                     0.00 
##        HousewifeHouseman                JobSearch                Pensioner 
##                     0.00                     0.00                     0.00 
##             NotSpecified 
##                     0.00

## [1] "Age"

##   vars   n  mean   sd min max range   se Q0.25 Q0.5 Q0.75
## 1    1 124 24.08 3.53  18  39    21 0.32    22 23.5    25

## [1] "Illness affecting sense of smell"

## 
##   Yes    No 
##  5.15 94.85

## [1] "montly net income"

##   vars  n   mean     sd min  max range    se Q0.25 Q0.5 Q0.75
## 1    1 97 798.97 396.63   0 2500  2500 40.27   600  750  1000

2.3 Tests for Homogeneity of Scent Subsamples

2.3.1 Metric variables

Age and monthly income

	peppermint	Vanilla	p	test
n	59	65
Age (mean (SD))	23.97 (3.41)	24.19 (3.65)	0.731
NetIncomeMonthly (mean (SD))	813.67 (357.68)	784.57 (434.67)	0.720

The Kruskal-Wallis-Test p-value for age is: 0.5556827 and that for Income is: 0.8649986 .

ANOVA Analysis

(#tab:unnamed-chunk-11) ANOVA of Age.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.12	1	122	12.54	.731	.001

Note. Note that tne IV here are the scent groups. (#tab:unnamed-chunk-11) ANOVA of NetIncomeMonthly.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.13	1	95	158,756.98	.720	.001

Note. Note that tne IV here are the scent groups.

Post hocs:

##  contrast             estimate    SE  df t.ratio p.value
##  peppermint - Vanilla    -0.22 0.637 122 -0.345  0.7306

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla     29.1 80.9 95 0.360   0.7200

---

2.3.2 Categorial variables

Gender, Federal state, ScentSensitivity, Job status, Illness affecting sense of smell

	peppermint	Vanilla	p	test
n	59	65
Gender (%)			0.536
Male	32 (54.2)	38 ( 58.5)
Female	27 (45.8)	26 ( 40.0)
Diverse	0 ( 0.0)	1 ( 1.5)
CountryNames (%)			0.921
BW	1 ( 1.7)	1 ( 1.5)
BY	4 ( 6.8)	3 ( 4.6)
BE	2 ( 3.4)	6 ( 9.2)
BB	2 ( 3.4)	2 ( 3.1)
HB	0 ( 0.0)	1 ( 1.5)
HH	1 ( 1.7)	0 ( 0.0)
HE	2 ( 3.4)	2 ( 3.1)
MV	2 ( 3.4)	1 ( 1.5)
NI	15 (25.4)	12 ( 18.5)
NW	4 ( 6.8)	3 ( 4.6)
RP	2 ( 3.4)	1 ( 1.5)
SL	0 ( 0.0)	1 ( 1.5)
SN	4 ( 6.8)	5 ( 7.7)
ST	16 (27.1)	21 ( 32.3)
SH	2 ( 3.4)	4 ( 6.2)
TH	1 ( 1.7)	2 ( 3.1)
other	1 ( 1.7)	0 ( 0.0)
ScentSensitivity (%)			NaN
Increased	11 (22.9)	12 ( 24.5)
Normal	28 (58.3)	31 ( 63.3)
Decreased	9 (18.8)	6 ( 12.2)
Impaired	0 ( 0.0)	0 ( 0.0)
JobStatus (%)			NaN
StudyPupilApprenticeship	57 (96.6)	60 ( 92.3)
Employee	2 ( 3.4)	5 ( 7.7)
SelfEmployedFreelance	0 ( 0.0)	0 ( 0.0)
HousewifeHouseman	0 ( 0.0)	0 ( 0.0)
JobSearch	0 ( 0.0)	0 ( 0.0)
Pensioner	0 ( 0.0)	0 ( 0.0)
NotSpecified	0 ( 0.0)	0 ( 0.0)
Illness = No (%)	43 (89.6)	49 (100.0)	0.063
Smoking = no (%)	41 (69.5)	54 ( 83.1)	0.116
Effort = LowEffort (%)	1 ( 2.1)	1 ( 2.0)	1.000

---

3 Scale reliability analysis

3.1 Mulidimensional Mood Questionnaire

Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. https://doi.org/10.1038/nature03701

Lichters, M., Brunnlieb, C., Nave, G., Sarstedt, M., & Vogt, B. (2016). The Influence of Serotonin Deficiency on Choice Deferral and the Compromise Effect. Journal of Marketing Research, 53(2), 183–198. https://doi.org/10.1509/jmr.14.0482

3.1.1 Pleasantness Dimension

Assessment of internal consistency reliability and convergent validity.

For reliability analyses and item statistics, we use the alpha() function from the psych package. Next to Cronbach’s alpha itself, the following helpful information is returned:

• G6 (Guttman’s lamba 6 relability is the average inter-item correlation)
• Reliability if item is dropped (>> are there raw_alpha values greater than overall alpha when every item is included? If so, this would indicate that the item should be deleted to increase the realibility of the scale)
• Item statistics: value in first columm labelled raw.r are the correlations between each item and the total score from the questionnaire (i.e., the item-total correlation)

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = MDMQ_p)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean  sd median_r
##       0.91      0.91    0.91      0.56  10 0.012    4 0.7     0.54
## 
##  lower alpha upper     95% confidence boundaries
## 0.88 0.91 0.93 
## 
##  Reliability if an item is dropped:
##                    raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r
## MDMQ_satisfied          0.90      0.90    0.90      0.56 8.9    0.014 0.0045
## MDMQ_bad                0.90      0.90    0.90      0.57 9.4    0.014 0.0051
## MDMQ_good               0.89      0.90    0.89      0.55 8.7    0.014 0.0041
## MDMQ_uncomfortable      0.90      0.90    0.90      0.56 9.1    0.014 0.0057
## MDMQ_comfortable        0.89      0.89    0.89      0.55 8.5    0.015 0.0046
## MDMQ_unhappy            0.89      0.90    0.89      0.56 8.7    0.014 0.0060
## MDMQ_unsatisfied        0.90      0.90    0.90      0.57 9.3    0.013 0.0049
## MDMQ_happy              0.89      0.89    0.89      0.55 8.4    0.015 0.0033
##                    med.r
## MDMQ_satisfied      0.54
## MDMQ_bad            0.56
## MDMQ_good           0.54
## MDMQ_uncomfortable  0.55
## MDMQ_comfortable    0.54
## MDMQ_unhappy        0.53
## MDMQ_unsatisfied    0.55
## MDMQ_happy          0.54
## 
##  Item statistics 
##                      n raw.r std.r r.cor r.drop mean   sd
## MDMQ_satisfied     124  0.79  0.78  0.74   0.71  3.6 0.94
## MDMQ_bad           124  0.72  0.74  0.69   0.65  4.5 0.67
## MDMQ_good          124  0.80  0.80  0.78   0.73  3.9 0.89
## MDMQ_uncomfortable 124  0.76  0.76  0.72   0.68  4.4 0.89
## MDMQ_comfortable   124  0.82  0.82  0.79   0.75  3.7 0.92
## MDMQ_unhappy       124  0.80  0.80  0.76   0.72  4.3 0.90
## MDMQ_unsatisfied   124  0.75  0.74  0.70   0.66  4.2 1.03
## MDMQ_happy         124  0.83  0.83  0.81   0.76  3.6 0.90
## 
## Non missing response frequency for each item
##                       1    2    3    4    5 miss
## MDMQ_satisfied     0.02 0.12 0.20 0.52 0.13    0
## MDMQ_bad           0.00 0.01 0.07 0.36 0.56    0
## MDMQ_good          0.02 0.06 0.16 0.53 0.23    0
## MDMQ_uncomfortable 0.01 0.04 0.10 0.24 0.60    0
## MDMQ_comfortable   0.01 0.10 0.25 0.45 0.19    0
## MDMQ_unhappy       0.00 0.07 0.07 0.30 0.56    0
## MDMQ_unsatisfied   0.02 0.07 0.08 0.31 0.51    0
## MDMQ_happy         0.02 0.10 0.27 0.48 0.14    0

##        Pleasantness     total
## alpha     0.9082951 0.9082951
## omega     0.9099398 0.9099398
## omega2    0.9099398 0.9099398
## omega3    0.9077040 0.9077040
## avevar    0.5622776 0.5622776

3.1.2 Wakefulness Dimension

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = MDMQ_w)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean   sd median_r
##        0.9       0.9    0.91      0.54 9.3 0.013  3.5 0.81     0.53
## 
##  lower alpha upper     95% confidence boundaries
## 0.88 0.9 0.93 
## 
##  Reliability if an item is dropped:
##                raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
## MDMQ_rested         0.89      0.89    0.90      0.53 8.0    0.015 0.0106  0.53
## MDMQ_weak           0.89      0.89    0.90      0.54 8.3    0.015 0.0085  0.53
## MDMQ_tired          0.89      0.89    0.89      0.54 8.1    0.015 0.0081  0.53
## MDMQ_lively         0.89      0.89    0.90      0.54 8.2    0.015 0.0107  0.53
## MDMQ_sleepy         0.88      0.88    0.88      0.51 7.4    0.017 0.0076  0.51
## MDMQ_awake          0.89      0.89    0.89      0.53 7.8    0.016 0.0084  0.52
## MDMQ_fresh          0.90      0.90    0.91      0.55 8.7    0.014 0.0098  0.53
## MDMQ_exhausted      0.89      0.90    0.90      0.55 8.5    0.015 0.0089  0.53
## 
##  Item statistics 
##                  n raw.r std.r r.cor r.drop mean   sd
## MDMQ_rested    124  0.79  0.79  0.75   0.71  3.1 1.13
## MDMQ_weak      124  0.76  0.75  0.72   0.67  3.7 1.04
## MDMQ_tired     124  0.78  0.78  0.75   0.70  3.5 1.07
## MDMQ_lively    124  0.75  0.76  0.71   0.68  3.3 0.98
## MDMQ_sleepy    124  0.85  0.85  0.84   0.79  3.7 1.08
## MDMQ_awake     124  0.80  0.81  0.78   0.74  3.5 0.95
## MDMQ_fresh     124  0.71  0.71  0.65   0.62  3.2 1.05
## MDMQ_exhausted 124  0.73  0.73  0.68   0.64  3.8 1.06
## 
## Non missing response frequency for each item
##                   1    2    3    4    5 miss
## MDMQ_rested    0.09 0.21 0.30 0.30 0.10    0
## MDMQ_weak      0.02 0.15 0.18 0.41 0.25    0
## MDMQ_tired     0.03 0.17 0.24 0.38 0.18    0
## MDMQ_lively    0.04 0.15 0.32 0.40 0.09    0
## MDMQ_sleepy    0.01 0.18 0.19 0.35 0.27    0
## MDMQ_awake     0.02 0.19 0.19 0.52 0.09    0
## MDMQ_fresh     0.04 0.24 0.27 0.35 0.10    0
## MDMQ_exhausted 0.00 0.12 0.29 0.23 0.36    0

##        Wakefulness     total
## alpha    0.9020401 0.9020401
## omega    0.9029420 0.9029420
## omega2   0.9029420 0.9029420
## omega3   0.9009454 0.9009454
## avevar   0.5404058 0.5404058

3.1.3 Calmness Dimension

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = MDMQ_c)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean   sd median_r
##       0.91      0.91    0.92      0.57  11 0.012  3.9 0.83     0.54
## 
##  lower alpha upper     95% confidence boundaries
## 0.89 0.91 0.94 
## 
##  Reliability if an item is dropped:
##               raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
## MDMQ_restless       0.9       0.9    0.91      0.58 9.5    0.013 0.0079  0.53
## MDMQ_serene         0.9       0.9    0.91      0.57 9.4    0.013 0.0077  0.56
## MDMQ_uneasy         0.9       0.9    0.91      0.57 9.3    0.014 0.0087  0.56
## MDMQ_relaxed        0.9       0.9    0.90      0.56 9.0    0.014 0.0073  0.52
## MDMQ_balanced       0.9       0.9    0.91      0.56 9.0    0.014 0.0078  0.56
## MDMQ_tense          0.9       0.9    0.91      0.57 9.3    0.014 0.0091  0.53
## MDMQ_nervous        0.9       0.9    0.91      0.58 9.5    0.013 0.0089  0.56
## MDMQ_calm           0.9       0.9    0.91      0.57 9.4    0.013 0.0072  0.56
## 
##  Item statistics 
##                 n raw.r std.r r.cor r.drop mean   sd
## MDMQ_restless 124  0.78  0.77  0.73   0.69  3.9 1.20
## MDMQ_serene   124  0.78  0.78  0.75   0.71  3.7 1.02
## MDMQ_uneasy   124  0.79  0.79  0.76   0.72  4.1 1.08
## MDMQ_relaxed  124  0.82  0.82  0.80   0.75  3.5 1.05
## MDMQ_balanced 124  0.81  0.81  0.79   0.74  3.5 1.05
## MDMQ_tense    124  0.80  0.79  0.76   0.72  4.1 1.08
## MDMQ_nervous  124  0.76  0.77  0.73   0.70  4.5 0.88
## MDMQ_calm     124  0.78  0.78  0.75   0.71  3.7 1.00
## 
## Non missing response frequency for each item
##                  1    2    3    4    5 miss
## MDMQ_restless 0.05 0.11 0.15 0.27 0.41    0
## MDMQ_serene   0.04 0.07 0.25 0.42 0.22    0
## MDMQ_uneasy   0.02 0.09 0.13 0.31 0.45    0
## MDMQ_relaxed  0.04 0.14 0.25 0.41 0.16    0
## MDMQ_balanced 0.02 0.18 0.26 0.37 0.17    0
## MDMQ_tense    0.02 0.10 0.11 0.31 0.45    0
## MDMQ_nervous  0.02 0.02 0.09 0.23 0.65    0
## MDMQ_calm     0.01 0.16 0.17 0.46 0.20    0

##         Calmness     total
## alpha  0.9126016 0.9126016
## omega  0.9129147 0.9129147
## omega2 0.9129147 0.9129147
## omega3 0.9099981 0.9099981
## avevar 0.5691390 0.5691390

3.1.4 Discriminant Validity according to heterotrait-monotrait ratio of correlations

Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A New Criterion for Assessing Discriminant Validity in Variance-based Structural Equation Modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8

HTMT Criterion

	Pleasantness	Wakefulness	Calmness
Pleasantness	1.0000000	0.6658641	0.8335743
Wakefulness	0.6658641	1.0000000	0.5208783
Calmness	0.8335743	0.5208783	1.0000000

3.2 Scent Arousal Scale

Bosmans, A. (2006). Scents and Sensibility: When Do (In)Congruent Ambient Scents Influence Product Evaluations? Journal of Marketing, 70(3), 32–43. https://doi.org/10.1509/jmkg.70.3.32

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = Arousal_Scale_Vars, check.keys = TRUE)
## 
##   raw_alpha std.alpha G6(smc) average_r  S/N   ase mean   sd median_r
##       0.44      0.47    0.45      0.18 0.88 0.083 0.93 0.87     0.12
## 
##  lower alpha upper     95% confidence boundaries
## 0.28 0.44 0.6 
## 
##  Reliability if an item is dropped:
##                             raw_alpha std.alpha G6(smc) average_r  S/N alpha se
## Arousal_RelaxedVSTense-          0.50      0.51    0.46     0.258 1.04    0.078
## Arousal_BoringVSStimulating      0.33      0.35    0.28     0.152 0.54    0.105
## Arousal_UnlivelyVSLively         0.24      0.24    0.18     0.097 0.32    0.115
## Arousal_DullVSBright             0.41      0.45    0.40     0.215 0.82    0.095
##                              var.r med.r
## Arousal_RelaxedVSTense-     0.0380 0.294
## Arousal_BoringVSStimulating 0.0154 0.096
## Arousal_UnlivelyVSLively    0.0025 0.096
## Arousal_DullVSBright        0.0372 0.147
## 
##  Item statistics 
##                              n raw.r std.r r.cor r.drop mean  sd
## Arousal_RelaxedVSTense-     90  0.60  0.53  0.20   0.14 0.88 1.7
## Arousal_BoringVSStimulating 90  0.59  0.66  0.50   0.30 1.03 1.2
## Arousal_UnlivelyVSLively    90  0.68  0.72  0.63   0.38 0.90 1.3
## Arousal_DullVSBright        90  0.60  0.58  0.32   0.21 0.92 1.5
## 
## Non missing response frequency for each item
##                               -3   -2   -1    0    1    2    3 miss
## Arousal_RelaxedVSTense      0.12 0.37 0.18 0.10 0.10 0.10 0.03 0.27
## Arousal_BoringVSStimulating 0.01 0.04 0.03 0.11 0.47 0.28 0.06 0.27
## Arousal_UnlivelyVSLively    0.01 0.07 0.07 0.18 0.27 0.37 0.04 0.27
## Arousal_DullVSBright        0.03 0.03 0.10 0.16 0.29 0.27 0.12 0.27

##          Arousal     total
## alpha  0.1863647 0.1863647
## omega  0.6606215 0.6606215
## omega2 0.6606215 0.6606215
## omega3 0.7529397 0.7529397
## avevar 0.5503456 0.5503456

3.3 Scent Pleasantness Scale

Bosmans, A. (2006). Scents and Sensibility: When Do (In)Congruent Ambient Scents Influence Product Evaluations? Journal of Marketing, 70(3), 32–43. https://doi.org/10.1509/jmkg.70.3.32

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = Pleasantness_Scale_Vars)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean  sd median_r
##       0.93      0.93    0.92      0.72  13 0.011 0.83 1.4      0.7
## 
##  lower alpha upper     95% confidence boundaries
## 0.91 0.93 0.95 
## 
##  Reliability if an item is dropped:
##                                         raw_alpha std.alpha G6(smc) average_r
## Pleasantness_BadVSGood                       0.90      0.90    0.88      0.70
## Pleasantness_UnpleasurableVSPleasurable      0.91      0.91    0.89      0.71
## Pleasantness_UncomfortableVSComfortable      0.92      0.92    0.91      0.74
## Pleasantness_NegativeVSPositive              0.90      0.90    0.88      0.69
## Pleasantness_UnattractiveVSAttractive        0.92      0.92    0.90      0.74
##                                          S/N alpha se  var.r med.r
## Pleasantness_BadVSGood                   9.2    0.014 0.0025  0.70
## Pleasantness_UnpleasurableVSPleasurable  9.8    0.014 0.0046  0.68
## Pleasantness_UncomfortableVSComfortable 11.6    0.012 0.0051  0.75
## Pleasantness_NegativeVSPositive          8.9    0.015 0.0039  0.67
## Pleasantness_UnattractiveVSAttractive   11.5    0.012 0.0040  0.73
## 
##  Item statistics 
##                                          n raw.r std.r r.cor r.drop mean  sd
## Pleasantness_BadVSGood                  90  0.90  0.91  0.89   0.85 1.04 1.5
## Pleasantness_UnpleasurableVSPleasurable 90  0.89  0.89  0.86   0.82 0.87 1.7
## Pleasantness_UncomfortableVSComfortable 90  0.85  0.84  0.78   0.75 0.64 1.6
## Pleasantness_NegativeVSPositive         90  0.91  0.91  0.90   0.86 1.02 1.6
## Pleasantness_UnattractiveVSAttractive   90  0.84  0.84  0.79   0.76 0.56 1.5
## 
## Non missing response frequency for each item
##                                           -3   -2   -1    0    1    2    3 miss
## Pleasantness_BadVSGood                  0.03 0.06 0.06 0.12 0.26 0.38 0.10 0.27
## Pleasantness_UnpleasurableVSPleasurable 0.08 0.02 0.12 0.08 0.27 0.30 0.13 0.27
## Pleasantness_UncomfortableVSComfortable 0.04 0.10 0.08 0.17 0.26 0.27 0.09 0.27
## Pleasantness_NegativeVSPositive         0.04 0.08 0.03 0.10 0.30 0.29 0.16 0.27
## Pleasantness_UnattractiveVSAttractive   0.07 0.06 0.07 0.21 0.32 0.22 0.06 0.27

##        Pleasantness     total
## alpha     0.9259173 0.9259173
## omega     0.9268759 0.9268759
## omega2    0.9268759 0.9268759
## omega3    0.9255454 0.9255454
## avevar    0.7183621 0.7183621

---

3.4 Perceived Social Density Scale

Madzharov, A. V., Block, L. G., & Morrin, M. (2015). The Cool Scent of Power: Effects of Ambient Scent on Consumer Preferences and Choice Behavior. Journal of Marketing, 79(1), 83–96. https://doi.org/10.1509/jm.13.0263

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = Social_Density_Scale_Vars)
## 
##   raw_alpha std.alpha G6(smc) average_r  S/N  ase mean  sd median_r
##        0.2       0.2    0.11      0.11 0.24 0.14    3 1.1     0.11
## 
##  lower alpha upper     95% confidence boundaries
## -0.09 0.2 0.48 
## 
##  Reliability if an item is dropped:
##                             raw_alpha std.alpha G6(smc) average_r S/N alpha se
## Social_Density_FullOfPeople     0.109      0.11   0.012      0.11  NA       NA
## Social_Density_Tight            0.012      0.11      NA        NA  NA       NA
##                             var.r med.r
## Social_Density_FullOfPeople 0.109  0.11
## Social_Density_Tight        0.012  0.11
## 
##  Item statistics 
##                              n raw.r std.r r.cor r.drop mean  sd
## Social_Density_FullOfPeople 97  0.71  0.74  0.25   0.11  3.1 1.4
## Social_Density_Tight        97  0.78  0.74  0.25   0.11  2.9 1.6
## 
## Non missing response frequency for each item
##                                1    2    3    4    5    6    7 miss
## Social_Density_FullOfPeople 0.08 0.36 0.21 0.12 0.19 0.03 0.01 0.22
## Social_Density_Tight        0.16 0.33 0.20 0.13 0.07 0.08 0.02 0.22

##        Social_Density     total
## alpha       0.1956596 0.1956596
## omega       0.2370247 0.2370247
## omega2      0.2370247 0.2370247
## omega3      0.2370247 0.2370247
## avevar      0.1542889 0.1542889

---

---

3.5 Perceived Complexity

Own creation

Herrmann, A., Zidansek, M., Sprott, D. E., & Spangenberg, E. R. (2013). The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales. Journal of Retailing, 89(1), 30–43. https://doi.org/10.1016/j.jretai.2012.08.002

Lévy, C. M., MacRae, A., & Köster, E. P. (2006). Perceived stimulus complexity and food preference development. Acta Psychologica, 123(3), 394–413. https://doi.org/10.1016/j.actpsy.2006.06.006

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = Simple_Complex_scale_Scale_Vars)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean sd median_r
##       0.68      0.68    0.66      0.42 2.1 0.052   42 17     0.51
## 
##  lower alpha upper     95% confidence boundaries
## 0.58 0.68 0.78 
## 
##  Reliability if an item is dropped:
##                            raw_alpha std.alpha G6(smc) average_r  S/N alpha se
## SimpleVSComplex                 0.29      0.29    0.17      0.17 0.42    0.126
## UncomplicatedVSComplicated      0.67      0.67    0.51      0.51 2.05    0.059
## PureVSDifferentiated            0.72      0.72    0.57      0.57 2.61    0.050
##                            var.r med.r
## SimpleVSComplex               NA  0.17
## UncomplicatedVSComplicated    NA  0.51
## PureVSDifferentiated          NA  0.57
## 
##  Item statistics 
##                             n raw.r std.r r.cor r.drop mean sd
## SimpleVSComplex            90  0.88  0.88  0.83   0.70   43 22
## UncomplicatedVSComplicated 90  0.73  0.74  0.57   0.42   41 22
## PureVSDifferentiated       90  0.73  0.72  0.50   0.38   42 24

##        Simple_Complex_scale       total
## alpha             0.6765754   0.6765754
## omega           159.1729757 159.1729757
## omega2          159.1729757 159.1729757
## omega3          116.4733557 116.4733557
## avevar          205.4379846 205.4379846

---

3.6 Kimchi and Palmer tasks

Kimchi, R., & Palmer, S. E. (1982). Form and texture in hierarchically constructed patterns. Journal of Experimental Psychology: Human Perception and Performance, 8(4), 521–535. https://doi.org/10.1037/0096-1523.8.4.521

## 
## Reliability analysis   
## Call: psych::alpha(x = Kimchi_Palmer_Scale_Vars)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N    ase mean sd median_r
##       0.93      0.93    0.94      0.54  14 0.0091   59 29     0.53
## 
##  lower alpha upper     95% confidence boundaries
## 0.91 0.93 0.95 
## 
##  Reliability if an item is dropped:
##     raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
## V1       0.93      0.93    0.94      0.55  14   0.0092 0.0063  0.56
## V2       0.92      0.93    0.93      0.53  12   0.0101 0.0087  0.53
## V3       0.92      0.92    0.93      0.53  12   0.0102 0.0084  0.52
## V4       0.92      0.92    0.93      0.53  12   0.0102 0.0083  0.52
## V5       0.93      0.93    0.94      0.55  13   0.0095 0.0083  0.55
## V6       0.93      0.93    0.94      0.54  13   0.0097 0.0078  0.55
## V7       0.93      0.93    0.94      0.53  13   0.0100 0.0089  0.52
## V8       0.93      0.93    0.93      0.53  13   0.0099 0.0075  0.53
## V9       0.93      0.93    0.94      0.54  13   0.0098 0.0090  0.55
## V10      0.93      0.93    0.94      0.53  13   0.0099 0.0087  0.53
## V11      0.92      0.93    0.93      0.53  12   0.0102 0.0082  0.53
## V12      0.93      0.93    0.93      0.53  12   0.0100 0.0079  0.53
## 
##  Item statistics 
##      n raw.r std.r r.cor r.drop mean sd
## V1  97  0.65  0.64  0.60   0.57   50 40
## V2  97  0.79  0.79  0.77   0.75   49 40
## V3  97  0.81  0.80  0.79   0.76   51 40
## V4  97  0.81  0.81  0.80   0.77   63 38
## V5  97  0.70  0.69  0.65   0.63   55 40
## V6  97  0.72  0.72  0.70   0.66   63 37
## V7  97  0.78  0.78  0.76   0.73   60 39
## V8  97  0.76  0.76  0.75   0.71   70 34
## V9  97  0.74  0.74  0.71   0.68   53 40
## V10 97  0.76  0.77  0.74   0.71   67 36
## V11 97  0.80  0.80  0.78   0.75   60 38
## V12 97  0.78  0.79  0.77   0.74   70 35

##        Kimchi_Palmer_Scale     total
## alpha            0.9318784 0.9318784
## omega            0.9323435 0.9323435
## omega2           0.9323435 0.9323435
## omega3           0.9314013 0.9314013
## avevar           0.5363625 0.5363625

4 Test Cool vs. Warm Scents

Story: Madzharov, A. V., Block, L. G., & Morrin, M. (2015). The Cool Scent of Power: Effects of Ambient Scent on Consumer Preferences and Choice Behavior. Journal of Marketing, 79(1), 83–96. https://doi.org/10.1509/jm.13.0263

Adams, C., & Doucé, L. (2017). What’s in a scent?: Meaning, shape, and sensorial concepts elicited by scents. Journal of Sensory Studies, 32(2), e12256. https://doi.org/10.1111/joss.12256

4.1 First: Cool vs. Warm, Madzharov et al. (2015)

(#tab:unnamed-chunk-23) Descriptives on warm vs. cool

ScentNumber	Mean	SD	Min	Max	sample
peppermint	2.13	1.19	1.00	6.00	59
Vanilla	4.09	1.78	1.00	7.00	65

ANOVA Analysis

(#tab:unnamed-chunk-24) ANOVA of Cool vs. Warm.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	38.25	1	88	2.26	< .001	.303

Note. Note that tne IV here are the scent groups.

Post hocs:

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    -1.96 0.317 88 -6.185  <.0001

4.2 Second: Cold vs. Hot, Adams & Doucé (2017)

(#tab:unnamed-chunk-26) Descriptives on hot vs. cold

ScentNumber	Mean	SD	Min	Max	sample
peppermint	23.85	19.50	0.00	68.00	59
Vanilla	44.52	19.01	0.00	77.00	65

ANOVA Analysis

(#tab:unnamed-chunk-27) ANOVA of Cold vs. Hot.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	25.90	1	88	371.10	< .001	.227

Note. Note that tne IV here are the scent groups.

Post hocs

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla    -20.7 4.06 88 -5.090  <.0001

---

4.3 Third perceived Social Density and perceived Threats in Germany

Problem is that Madzharov et al.’s (2015) scale for perceived social density is not internally consisitent. We, therefore, analyze both items seperately.

---

4.3.0.1 Does it seem there are a lot of people around you right now?

---

(1: not at all | 7: strongly agree)

ANOVA Analysis

(#tab:unnamed-chunk-29) ANOVA of Cold vs. Hot.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	1.02	1	95	2.01	.315	.011

Note. Note that tne IV here are the scent groups.

Post hocs

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    0.291 0.288 95 1.010   0.3150

---

4.3.0.2 How spacious do you think this room is?

---

(1: not at all | 7: very)

ANOVA Analysis

(#tab:unnamed-chunk-31) ANOVA of Cold vs. Hot.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.10	1	95	2.53	.753	.001

Note. Note that tne IV here are the scent groups.

Post hocs

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    0.102 0.323 95 0.316   0.7528

---

4.3.0.3 New Quesion Extent of distraction from others while working on the study?

---

(1: not at all | 7: very)

ANOVA Analysis

(#tab:unnamed-chunk-33) ANOVA of Cold vs. Hot.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	8.62	1	95	1.29	.004	.083

Note. Note that tne IV here are the scent groups.

Post hocs

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    0.679 0.231 95 2.937   0.0042

---

4.3.0.4 New Quesion Level of perceived Safety in living in Germany?

---

(1: not at all | 7: very)

ANOVA Analysis

(#tab:unnamed-chunk-35) ANOVA of Cold vs. Hot.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.21	1	95	1.23	.650	.002

Note. Note that tne IV here are the scent groups.

Post hocs

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    0.102 0.225 95 0.455   0.6499

5 Test simple vs. complex Scents

Story: Herrmann, A., Zidansek, M., Sprott, D. E., & Spangenberg, E. R. (2013). The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales. Journal of Retailing, 89(1), 30–43. https://doi.org/10.1016/j.jretai.2012.08.002

5.1 First: Complexity item (Lévy et al., 2006)

Lévy, C. M., MacRae, A., & Köster, E. P. (2006). Perceived stimulus complexity and food preference development. Acta Psychologica, 123(3), 394–413. https://doi.org/10.1016/j.actpsy.2006.06.006

ANOVA Analysis

(#tab:unnamed-chunk-37) ANOVA of Complexity.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	4.82	1	88	458.07	.031	.052

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla    -9.91 4.51 88 -2.196  0.0307

5.2 Second : 2 new items by Herrmann et al. (2013): Uncomplicated vs. Complicated | Pure vs. Differentiated

Herrmann, A., Zidansek, M., Sprott, D. E., & Spangenberg, E. R. (2013). The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales. Journal of Retailing, 89(1), 30–43. https://doi.org/10.1016/j.jretai.2012.08.002

5.2.1 Uncomplicated vs. Complicated

ANOVA Analysis

(#tab:unnamed-chunk-39) ANOVA of Complexity.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.00	1	88	475.51	.962	.000

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate  SE df t.ratio p.value
##  peppermint - Vanilla    -0.22 4.6 88 -0.048  0.9619

5.2.2 Pure vs. Differentiated

ANOVA Analysis

(#tab:unnamed-chunk-41) ANOVA of Complexity.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	13.84	1	88	486.29	< .001	.136

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla    -17.3 4.65 88 -3.720  0.0003

---

5.3 Third: Multi-item Scale consisting of all above-analyzed items

ANOVA Analysis

(#tab:unnamed-chunk-43) ANOVA of Simple_Complex_scale.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	6.56	1	88	286.59	.012	.069

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla    -9.14 3.57 88 -2.562  0.0121

5.4 Fourth: Retrieval Fluency of scent-related Thoughts (Graf et al., 2018)

Idea: Complex vs. simple scent should go hand in hand with more disfluent retrieval

Herrmann, A., Zidansek, M., Sprott, D. E., & Spangenberg, E. R. (2013). The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales. Journal of Retailing, 89(1), 30–43. https://doi.org/10.1016/j.jretai.2012.08.002

Graf, L. K. M., Mayer, S., & Landwehr, J. R. (2018). Measuring Processing Fluency: One versus Five Items. Journal of Consumer Psychology, 28(3), 393–411. https://doi.org/10.1002/jcpy.1021

ANOVA Analysis

(#tab:unnamed-chunk-45) ANOVA of Complexity.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	6.20	1	88	967.35	.015	.066

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate   SE df t.ratio p.value
##  peppermint - Vanilla     16.3 6.56 88 2.490   0.0146

6 Published pretests on characteristics that should not differ between scents

6.1 Pleasantness of the scent

Bosmans, A. (2006). Scents and Sensibility: When Do (In)Congruent Ambient Scents Influence Product Evaluations? Journal of Marketing, 70(3), 32–43. https://doi.org/10.1509/jmkg.70.3.32

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

ANOVA Analysis

(#tab:unnamed-chunk-47) ANOVA of Scent Pleasantness.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	2.25	1	88	1.91	.137	.025

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla   -0.437 0.291 88 -1.500  0.1373

---

---

6.2 Arousal of the scent

Bosmans, A. (2006). Scents and Sensibility: When Do (In)Congruent Ambient Scents Influence Product Evaluations? Journal of Marketing, 70(3), 32–43. https://doi.org/10.1509/jmkg.70.3.32

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

ANOVA Analysis

(#tab:unnamed-chunk-49) ANOVA of Scent_Arousal.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	7.33	1	88	0.55	.008	.077

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla    0.423 0.156 88 2.708   0.0081

---

---

6.3 Familiarity of the scent

Morrin, M., & Ratneshwar, S. (2003). Does It Make Sense to Use Scents to Enhance Brand Memory? Journal of Marketing Research, 40(1), 10–25. https://doi.org/10.1509/jmkr.40.1.10.19128

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

ANOVA Analysis

(#tab:unnamed-chunk-51) ANOVA of Familiarity_r1.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	1.54	1	88	2.46	.218	.017

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla     0.41 0.331 88 1.240   0.2181

---

6.4 Intensity of the scent

Spangenberg, E. R., Crowley, A. E., & Henderson, P. W. (1996). Improving the Store Environment: Do Olfactory Cues Affect Evaluations and Behaviors? Journal of Marketing, 60(2), 67–80. https://doi.org/10.2307/1251931

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

ANOVA Analysis

(#tab:unnamed-chunk-53) ANOVA of Intensity_r1.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.00	1	88	1.25	.977	.000

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	NA
Vanilla	NA

##  contrast             estimate    SE df t.ratio p.value
##  peppermint - Vanilla  0.00692 0.236 88 0.029   0.9767

*******************************************************

6.5 Perceived Indoor Air Quality

Zhang, H., Arens, E., & Pasut, W. (2011). Air temperature thresholds for indoor comfort and perceived air quality. Building Research & Information, 39(2), 134–144. https://doi.org/10.1080/09613218.2011.552703

Girard, A., Lichters, M., Sarstedt, M., & Biswas, D. (2019). Short- and Long-term Effects of nonconsciously Processed Ambient Scents in a Servicescape: Findings from two Field Experiments. Journal of Service Research. Advance online publication. https://doi.org/10.1177/1094670519842333

6.5.1 Only consumers that did recognize the ambient scent and which sense of smell ist not comprised (illness or chronically)

ANOVA Analysis

(#tab:unnamed-chunk-55) ANOVA of OverallAirQuality_1.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.03	1	122	515.01	.854	.000

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	72.66102
Vanilla	73.41538

##  contrast             estimate   SE  df t.ratio p.value
##  peppermint - Vanilla   -0.754 4.08 122 -0.185  0.8536

---

6.6 Multidimensional Mood Questionnaire (MDMQ)

Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. https://doi.org/10.1038/nature03701

Lichters, M., Brunnlieb, C., Nave, G., Sarstedt, M., & Vogt, B. (2016). The Influence of Serotonin Deficiency on Choice Deferral and the Compromise Effect. Journal of Marketing Research, 53(2), 183–198. https://doi.org/10.1509/jmr.14.0482

6.6.1 First: Each single item

MDMQ single items

	peppermint	Vanilla	p	test
n	59	65
MDMQ_satisfied (mean (SD))	3.63 (0.69)	3.60 (1.13)	0.874
MDMQ_rested (mean (SD))	3.25 (1.08)	3.00 (1.17)	0.212
MDMQ_restless (mean (SD))	3.93 (1.20)	3.85 (1.21)	0.693
MDMQ_bad (mean (SD))	4.42 (0.67)	4.51 (0.66)	0.487
MDMQ_weak (mean (SD))	3.69 (1.04)	3.77 (1.06)	0.694
MDMQ_serene (mean (SD))	3.68 (1.09)	3.72 (0.96)	0.807
MDMQ_tired (mean (SD))	3.59 (0.95)	3.42 (1.17)	0.358
MDMQ_good (mean (SD))	3.93 (0.81)	3.85 (0.96)	0.591
MDMQ_uneasy (mean (SD))	4.03 (1.16)	4.11 (1.00)	0.704
MDMQ_lively (mean (SD))	3.46 (0.79)	3.23 (1.11)	0.199
MDMQ_uncomfortable (mean (SD))	4.41 (0.79)	4.38 (0.98)	0.891
MDMQ_relaxed (mean (SD))	3.69 (1.04)	3.35 (1.04)	0.070
MDMQ_sleepy (mean (SD))	3.83 (0.99)	3.58 (1.16)	0.208
MDMQ_comfortable (mean (SD))	3.68 (0.88)	3.72 (0.96)	0.786
MDMQ_balanced (mean (SD))	3.46 (1.04)	3.51 (1.06)	0.792
MDMQ_unhappy (mean (SD))	4.39 (0.85)	4.29 (0.95)	0.549
MDMQ_awake (mean (SD))	3.61 (0.81)	3.37 (1.05)	0.159
MDMQ_unsatisfied (mean (SD))	4.24 (0.97)	4.18 (1.09)	0.777
MDMQ_tense (mean (SD))	4.14 (1.15)	4.02 (1.02)	0.539
MDMQ_fresh (mean (SD))	3.46 (1.06)	3.00 (1.00)	0.015
MDMQ_happy (mean (SD))	3.71 (0.77)	3.55 (1.00)	0.329
MDMQ_nervous (mean (SD))	4.39 (0.85)	4.52 (0.90)	0.401
MDMQ_exhausted (mean (SD))	3.93 (0.98)	3.74 (1.12)	0.310
MDMQ_calm (mean (SD))	3.73 (1.03)	3.65 (0.98)	0.647

6.6.2 Second: Each of the 3 dimensions Pleasantness, Calmness, Wakefulness

6.6.2.1 Only consumers that did recognize the ambient scent and which sense of smell ist not comprised (illness or chronically)

Pleasantness

ANOVA Analysis

(#tab:unnamed-chunk-58) ANOVA of MDMQ_PleasantnessDim.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.10	1	122	0.49	.756	.001

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	4.050847
Vanilla	4.011539

##  contrast             estimate    SE  df t.ratio p.value
##  peppermint - Vanilla   0.0393 0.126 122 0.311   0.7562

---

---

Wakefulness

ANOVA Analysis

(#tab:unnamed-chunk-60) ANOVA of MDMQ_WakefulnessDim.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	2.23	1	122	0.64	.138	.018

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	3.603814
Vanilla	3.388462

##  contrast             estimate    SE  df t.ratio p.value
##  peppermint - Vanilla    0.215 0.144 122 1.493   0.1380

---

---

Calmness

ANOVA Analysis

(#tab:unnamed-chunk-62) ANOVA of MDMQ_CalmnessDim.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.08	1	122	0.69	.784	.001

Note. Note that tne IV here are the scent groups.

Post hocs:

Mean for each Scent

	x
peppermint	3.881356
Vanilla	3.840385

##  contrast             estimate    SE  df t.ratio p.value
##  peppermint - Vanilla    0.041 0.149 122 0.275   0.7838

---

6.7 Scent identification

Fishers exact test Scent Identification

	peppermint	Vanilla
No	44.1	53.8
Yes	55.9	46.2
Total	100.0	100.0
Count	59.0	65.0

## [1] "Fisher's undirected p value:"

## [1] 0.2875509

---

7 Published associations triggered by ambient scents that should not differ between scents

Adams, C., & Doucé, L. (2017). What’s in a scent?: Meaning, shape, and sensorial concepts elicited by scents. Journal of Sensory Studies, 32(2), e12256. https://doi.org/10.1111/joss.12256

Herrmann, A., Zidansek, M., Sprott, D. E., & Spangenberg, E. R. (2013). The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales. Journal of Retailing, 89(1), 30–43. https://doi.org/10.1016/j.jretai.2012.08.002

7.1 Basic associations

include: active vs. passive, weak vs. strong, feminine vs. masculine from Adams & Douce (2017), as well as simple vs. complex, uncomplicated vs. complicated, pure vs. differentiated from Herrmann et al. (2013), and additionally artificial vs. natural

BasicVAS single items

	peppermint	Vanilla	p	test
n	59	65
ActiveVSPassive (mean (SD))	36.15 (25.65)	40.36 (21.91)	0.405
WeakVSStrong (mean (SD))	66.22 (20.43)	65.61 (18.43)	0.883
FeminineVSMasculine (mean (SD))	50.61 (18.16)	33.39 (16.16)	<0.001
ArtificialVSNatural (mean (SD))	38.91 (25.11)	41.43 (22.88)	0.621
SimpleVSComplex (mean (SD))	38.57 (20.54)	48.48 (22.27)	0.031
UncomplicatedVSComplicated (mean (SD))	40.85 (21.93)	41.07 (21.68)	0.962
PureVSDifferentiated (mean (SD))	33.11 (22.26)	50.41 (21.83)	<0.001

---

7.2 Cross-modal associations

include: bright vs. dim, cold vs. hot, fragile vs. sturdy, high vs. low, light vs. dark, light vs. heavy, loud vs. quite, rough vs. smooth, shallow vs. deep, soft vs. hard.

CrossModalVAS single items

	peppermint	Vanilla	p	test
n	59	65
BrightVSDim (mean (SD))	26.87 (18.10)	43.05 (21.77)	<0.001
ColdVSHot (mean (SD))	23.85 (19.50)	44.52 (19.01)	<0.001
FragileVSSturdy (mean (SD))	52.54 (16.88)	47.70 (20.03)	0.218
HighVSLow (mean (SD))	42.28 (19.73)	47.05 (18.51)	0.241
LightVSDark (mean (SD))	31.96 (17.09)	37.61 (20.13)	0.154
LightVSHeavy (mean (SD))	37.09 (18.01)	48.25 (25.34)	0.018
LoudVSQuiet (mean (SD))	54.24 (20.08)	56.61 (19.31)	0.569
RoughVSSmooth (mean (SD))	60.57 (20.76)	68.91 (18.88)	0.049
ShallowVSDeep (mean (SD))	47.87 (19.03)	51.68 (21.62)	0.376
SoftVSHard (mean (SD))	52.67 (22.59)	30.14 (18.86)	<0.001

---

8 Room temperature perception and estimates

8.1 Perceived Room warmth

(#tab:unnamed-chunk-67) Descriptives on hot vs. cold

ScentNumber	Mean	SD	Min	Max	sample
peppermint	3.83	1.33	1.00	6.00	59
Vanilla	4.59	1.32	2.00	9.00	65

ANOVA Analysis

(#tab:unnamed-chunk-68) ANOVA of Perceived Temparture (1=Cold | 9=Warm).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	7.96	1	95	1.75	.006	.077

Note. Note that tne IV here are the scent groups.

Descriptives for each Scent

(#tab:unnamed-chunk-69) Descriptives on Actual Temperature for each scent

ScentNumber	Mean	SD	Min	Max
peppermint	24.43	0.31	24.20	24.95
Vanilla	24.75	0.36	24.30	25.25

---

8.2 Estimated Room Temperature in C°

ANOVA Analysis

(#tab:unnamed-chunk-70) ANOVA of Estimated Temparture (degree Celsius).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.06	1	95	4.18	.804	.001

Note. Note that tne IV here are the scent groups.

8.3 Deviation Estimated - Actual Room Temperature in Degree Celsius

8.3.1 Simple Error

ANOVA Analysis

(#tab:unnamed-chunk-71) ANOVA of Estimated-Actual Temparture (degree Celsius).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.12	1	95	4.21	.733	.001

Note. Note that tne IV here are the scent groups.

8.3.2 Squared Error

ANOVA Analysis

(#tab:unnamed-chunk-72) ANOVA of (Estimated-Actual)² Temparture (degree Celsius).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.35	1	95	167.39	.558	.004

Note. Note that tne IV here are the scent groups.

The rest on this script is devoted to assess effects in consumer decision making between vanilla and peppermint scent

9 Prestige-focused vs. comfort-focused ad for a car

Story: Study 1 in: Madzharov, A. V., Block, L. G., & Morrin, M. (2015). The Cool Scent of Power: Effects of Ambient Scent on Consumer Preferences and Choice Behavior. Journal of Marketing, 79(1), 83–96. https://doi.org/10.1509/jm.13.0263

9.1 Liking of ads

ANOVA Analysis

(#tab:unnamed-chunk-73) ANOVA of Ad Liking (1=prestige | 9=comfort).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.02	1	122	5.00	.887	.000

Note. Note that tne IV here are the scent groups.

9.2 Effectiveness of ads

ANOVA Analysis

(#tab:unnamed-chunk-74) ANOVA of Ad Liking (1=prestige | 9=comfort).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.06	1	122	6.30	.799	.001

Note. Note that tne IV here are the scent groups.

9.3 Combination of both items

First build scale values as mean of item scores.

Cronbach’s Alpha and G6, and AVE

## 
## Reliability analysis   
## Call: psych::alpha(x = Ad_Scale_Vars)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean sd median_r
##       0.56      0.56    0.39      0.39 1.3 0.079  6.8  2     0.39
## 
##  lower alpha upper     95% confidence boundaries
## 0.4 0.56 0.71 
## 
##  Reliability if an item is dropped:
##                       raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r
## CarAdPreference_r1         0.39      0.39    0.15      0.39  NA       NA  0.39
## CarAdEffectiveness_r1      0.15      0.39      NA        NA  NA       NA  0.15
##                       med.r
## CarAdPreference_r1     0.39
## CarAdEffectiveness_r1  0.39
## 
##  Item statistics 
##                         n raw.r std.r r.cor r.drop mean  sd
## CarAdPreference_r1    124  0.81  0.83  0.52   0.39  7.3 2.2
## CarAdEffectiveness_r1 124  0.85  0.83  0.52   0.39  6.3 2.5
## 
## Non missing response frequency for each item
##                          1    2    3    4    5    6    7    8    9 miss
## CarAdPreference_r1    0.05 0.02 0.04 0.02 0.03 0.02 0.21 0.24 0.36    0
## CarAdEffectiveness_r1 0.05 0.05 0.13 0.03 0.02 0.13 0.20 0.14 0.25    0

##               Ad     total
## alpha  0.5553194 0.5553194
## omega  0.5681149 0.5681149
## omega2 0.5681149 0.5681149
## omega3 0.5681149 0.5681149
## avevar 0.4021029 0.4021029

ANOVA Analysis

(#tab:unnamed-chunk-76) ANOVA of Ad Effects (1=prestige | 9=comfort).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.01	1	122	3.91	.936	.000

Note. Note that tne IV here are the scent groups.

10 Preference for luxury vs. nonluxery FMCG

Story: Ma, H., Bradshaw, H. K., Janakiraman, N., & Hill, S. E. (2019). Spending as protection: The need for safety increases preference for luxury products. Marketing Letters, 30(1), 45–56. https://doi.org/10.1007/s11002-019-09480-0

(#tab:unnamed-chunk-77) Descriptives on luxury vs. common brand

ScentNumber	Mean	SD	Min	Max	sample
peppermint	5.03	2.62	1.00	9.00	59
Vanilla	6.15	2.61	1.00	9.00	65

ANOVA Analysis

(#tab:unnamed-chunk-78) ANOVA of Coffee Preference (1=mundane | 9=luxury).

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	3.93	1	84	6.84	.051	.045

Note. Note that tne IV here are the scent groups.

[1] “one-sided t-test p:” [1] 0.02545431

11 Election of political parties

Fishers exact test Elections (counts)

	peppermint	Vanilla
CDU	2	3
SPD	2	2
DieGruenen	15	21
AFD	0	4
DieLinke	8	4
FDP	10	8
CSU	1	0
DieBlauePartei	0	0
FreieWaehler	0	0
BuergerInWut	0	0
LKR	0	1
NPD	0	0
OeDP	1	1
DiePartei	9	5

Fishers exact test Elections (%)

	peppermint	Vanilla
CDU	4.2	6.1
SPD	4.2	4.1
DieGruenen	31.2	42.9
AFD	0.0	8.2
DieLinke	16.7	8.2
FDP	20.8	16.3
CSU	2.1	0.0
DieBlauePartei	0.0	0.0
FreieWaehler	0.0	0.0
BuergerInWut	0.0	0.0
LKR	0.0	2.0
NPD	0.0	0.0
OeDP	2.1	2.0
DiePartei	18.8	10.2
Total	100.1	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.3088456

11.1 Right-wing vs. other parties voting

Fishers exact test Elections

	peppermint	Vanilla
others	100	89.8
Right-wing	0	10.2
Total	100	100.0
Count	48	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.05615844

## [1] "Fisher's directed p value (expectation: consumers under vanilla vote more for right-wing parties):"

## [1] 0.02958885

12 Purchase rates in grocceries

12.1 Pooled analysis across all product categories

Descriptives

(#tab:unnamed-chunk-81) Descriptives on Purchase rates for each scent

ScentNumber	Mean	SD	Min	Max
peppermint	0.77	0.24	0.25	1.00
Vanilla	0.75	0.25	0.00	1.00

ANOVA Analysis

(#tab:unnamed-chunk-82) ANOVA of Purchase rates across 4 purchase situations.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.18	1	95	0.06	.674	.002

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Product_Choice_PurchaseRate by ScentNumber
## W = 1225.5, p-value = 0.7076
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -6.860703e-06  2.519429e-05
## sample estimates:
## difference in location 
##           6.993813e-05

---

---

12.2 Category-wise analysis

12.2.1 Chocolate

Fishers exact test Purchase

	peppermint	Vanilla
Buy	79.2	83.7
No-Buy	20.8	16.3
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.610195

## [1] "Fisher's directed p value:"

## [1] 0.3785785

---

12.2.2 Orange Juice

Fishers exact test Purchase

	peppermint	Vanilla
Buy	87.5	85.7
No-Buy	12.5	14.3
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 1

## [1] "Fisher's directed p value:"

## [1] 0.7100059

---

12.2.3 Wine

Fishers exact test Purchase

	peppermint	Vanilla
Buy	66.7	65.3
No-Buy	33.3	34.7
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 1

## [1] "Fisher's directed p value:"

## [1] 0.6387398

---

12.2.4 Chips

Fishers exact test Purchase

	peppermint	Vanilla
Buy	75	65.3
No-Buy	25	34.7
Total	100	100.0
Count	48	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.3762578

## [1] "Fisher's directed p value:"

## [1] 0.897184

---

13 Premium rates in grocceries across 4 purchase decisions

13.1 First relative premium rate (#premium/#total purchases)

(#tab:unnamed-chunk-88) Descriptives on Purchase rates for each scent

ScentNumber	Mean	SD	Min	Max
peppermint	0.42	0.33	0.00	1.00
Vanilla	0.46	0.33	0.00	1.00

ANOVA Analysis

(#tab:unnamed-chunk-89) ANOVA of Premium rates across 4 purchase situations.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.32	1	94	0.11	.571	.003

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Product_Choice_PremiumRate by ScentNumber
## W = 1058, p-value = 0.4862
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -1.666799e-01  3.911517e-05
## sample estimates:
## difference in location 
##          -1.766801e-05

---

13.2 Category-wise analysis

13.2.1 Chocolate

Fishers exact test Purchase

	peppermint	Vanilla
Common	50.0	44.9
Premium	29.2	38.8
No-Buy	20.8	16.3
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.4944873

13.2.2 Orange Juice

Fishers exact test Purchase

	peppermint	Vanilla
Common	43.8	36.7
Premium	43.8	49.0
No-Buy	12.5	14.3
Total	100.1	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.6619981

---

13.2.3 Wine

Fishers exact test Purchase

	peppermint	Vanilla
Common	43.8	49.0
Premium	22.9	16.3
No-Buy	33.3	34.7
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.5849733

---

13.2.4 Chips

Fishers exact test Purchase

	peppermint	Vanilla
Common	37.5	30.6
Premium	37.5	34.7
No-Buy	25.0	34.7
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.813033

---

13.3 Second: premium rate (#premium/#4)

(#tab:unnamed-chunk-95) Descriptives on Purchase rates for each scent

ScentNumber	Mean	SD	Min	Max
peppermint	0.33	0.26	0.00	1.00
Vanilla	0.35	0.27	0.00	0.75

ANOVA Analysis

(#tab:unnamed-chunk-96) ANOVA of Premium rates across 4 purchase situations.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.06	1	95	0.07	.802	.001

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Product_Choice_PremiumRate_baseAllDecisions by ScentNumber
## W = 1137.5, p-value = 0.7761
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -2.499727e-01  2.422877e-05
## sample estimates:
## difference in location 
##          -6.363314e-05

---

14 Color-associations triggered by ambient scents

Now, filter out participants that are colorblind

	peppermint	Vanilla	p	test
n	45	48
Color_Associations_red (mean (SD))	1.82 (1.43)	3.15 (1.84)	<0.001
Color_Associations_green (mean (SD))	4.80 (2.21)	2.94 (1.77)	<0.001
Color_Associations_pink (mean (SD))	2.20 (1.67)	3.42 (1.88)	0.001
Color_Associations_gray (mean (SD))	2.80 (2.03)	2.50 (1.96)	0.470
Color_Associations_orange (mean (SD))	1.76 (1.37)	3.65 (2.03)	<0.001
Color_Associations_blue (mean (SD))	3.69 (2.12)	2.88 (1.75)	0.046
Color_Associations_black (mean (SD))	1.93 (1.76)	1.75 (1.45)	0.584
Color_Associations_white (mean (SD))	5.56 (1.57)	4.21 (2.06)	0.001
Color_Associations_yellow (mean (SD))	2.56 (1.93)	4.65 (1.99)	<0.001
Color_Associations_purple (mean (SD))	2.36 (1.84)	3.06 (1.68)	0.056
Color_Associations_brown (mean (SD))	2.02 (1.70)	3.38 (2.29)	0.002

15 Brand Logo Liking

ANOVA Analysis

(#tab:unnamed-chunk-100) ANOVA of Logo Liking.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.08	1	91	633.51	.774	.001

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  LogoLiking_1 by ScentNumber
## W = 1137, p-value = 0.6639
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -8.999955 11.999999
## sample estimates:
## difference in location 
##               2.000031

---

16 Brand Logo triggered Associations

	peppermint	Vanilla	p	test
n	45	48
Logo_dynamic (mean (SD))	5.13 (1.63)	5.08 (1.49)	0.877
Logo_brave (mean (SD))	4.27 (1.64)	4.23 (1.59)	0.911
Logo_sentimental (mean (SD))	2.78 (1.54)	2.77 (1.77)	0.984
Logo_reliable (mean (SD))	3.04 (1.52)	3.21 (1.29)	0.576
Logo_active (mean (SD))	5.07 (1.42)	4.90 (1.37)	0.557
Logo_romantic (mean (SD))	2.98 (1.74)	2.77 (1.93)	0.589
Logo_aggrassive (mean (SD))	4.24 (1.52)	3.48 (1.69)	0.024
Logo_ordinary (mean (SD))	2.98 (1.59)	2.52 (1.29)	0.130
Logo_down_to_earth (mean (SD))	2.33 (1.21)	2.65 (1.41)	0.255
Logo_plain (mean (SD))	2.80 (1.65)	2.31 (1.57)	0.148
Logo_innovative (mean (SD))	3.76 (1.58)	3.88 (1.63)	0.721
Logo_responsibly (mean (SD))	2.96 (1.38)	3.25 (1.66)	0.356

17 Anagrams

(#tab:unnamed-chunk-103) Descriptives on correctly solved Anagrams

ScentNumber	Mean	SD	Min	Max	sample
peppermint	5.67	2.12	0.00	10.00	59
Vanilla	5.24	1.98	0.00	9.00	65

ANOVA Analysis

(#tab:unnamed-chunk-104) ANOVA of correctly solved Anagrams.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	1.03	1	95	4.21	.314	.011

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Anagramm_Total_correct by ScentNumber
## W = 1333, p-value = 0.2528
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -5.499321e-06  1.000050e+00
## sample estimates:
## difference in location 
##           8.308531e-05

---

---

17.1 Anagrams in isolation

Ampel

Fishers exact test

	peppermint	Vanilla
0	41.7	49
1	58.3	51
Total	100.0	100
Count	48.0	49

## [1] "Fisher's undirected p value:"

## [1] 0.5425965

Asche

Fishers exact test

	peppermint	Vanilla
0	60.4	67.3
1	39.6	32.7
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.5299101

Salat

Fishers exact test

	peppermint	Vanilla
0	87.5	89.8
1	12.5	10.2
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.759275

Eichel

Fishers exact test

	peppermint	Vanilla
0	31.2	40.8
1	68.8	59.2
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.3992745

Gras

Fishers exact test

	peppermint	Vanilla
0	35.4	36.7
1	64.6	63.3
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 1

Reifen

Fishers exact test

	peppermint	Vanilla
0	62.5	49
1	37.5	51
Total	100.0	100
Count	48.0	49

## [1] "Fisher's undirected p value:"

## [1] 0.2217692

Tor

Fishers exact test

	peppermint	Vanilla
0	4.2	2
1	95.8	98
Total	100.0	100
Count	48.0	49

## [1] "Fisher's undirected p value:"

## [1] 0.617146

Karl

Fishers exact test

	peppermint	Vanilla
0	75	79.6
1	25	20.4
Total	100	100.0
Count	48	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.6342137

Bier

Fishers exact test

	peppermint	Vanilla
0	25	44.9
1	75	55.1
Total	100	100.0
Count	48	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.05536255

Nebel

Fishers exact test

	peppermint	Vanilla
0	10.4	16.3
1	89.6	83.7
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.5529392

Combine Eichel and Gras as Anagrams with a solution making a potentail threat salient

Fishers exact test

	peppermint	Vanilla
0	14.6	18.4
1	37.5	40.8
2	47.9	40.8
Total	100.0	100.0
Count	48.0	49.0

## [1] "Fisher's undirected p value:"

## [1] 0.7668248

---

---

18 Analysis of open question on Scent Associations

Coding to categories was done manually

18.1 Total number of different scent associations

(#tab:unnamed-chunk-117) Descriptives: total number of different scent associations

ScentNumber	Mean	SD	Min	Max	sample
peppermint	2.54	1.36	0.00	7.00	59
Vanilla	2.32	1.01	0.00	5.00	65

ANOVA Analysis

(#tab:unnamed-chunk-118) ANOVA

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	0.79	1	88	1.44	.376	.009

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  OA_Total by ScentNumber
## W = 1067, p-value = 0.6457
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -9.555132e-06  9.999243e-01
## sample estimates:
## difference in location 
##           5.333706e-05

---

18.2 Each scent associations in isolation

	peppermint	Vanilla	p	test
n	59	65
OA_Floral = 1 (%)	0 ( 0.0)	4 ( 9.1)	0.114
OA_Fresh = 1 (%)	21 ( 45.7)	12 ( 27.3)	0.112
OA_Pleasant = 1 (%)	5 ( 10.9)	14 ( 31.8)	0.030
OA_Unpleasant = 1 (%)	2 ( 4.3)	2 ( 4.5)	1.000
OA_fruity = 1 (%)	1 ( 2.2)	1 ( 2.3)	1.000
OA_Vanilla = 1 (%)	0 ( 0.0)	17 ( 38.6)	<0.001
OA_Mint_Eucalyptus = 1 (%)	35 ( 76.1)	2 ( 4.5)	<0.001
OA_Household_Cleaner_Personal_Care = 1 (%)	0 ( 0.0)	2 ( 4.5)	0.455
OA_Tasty_hungry = 1 (%)	0 ( 0.0)	1 ( 2.3)	0.982
OA_too_intense = 1 (%)	4 ( 8.7)	1 ( 2.3)	0.385
OA_Coffee = 1 (%)	0 ( 0.0)	3 ( 6.8)	0.225
OA_Warm = 1 (%)	0 ( 0.0)	2 ( 4.5)	0.455
OA_Summer = 1 (%)	0 ( 0.0)	1 ( 2.3)	0.982
OA_Cold = 1 (%)	10 ( 21.7)	3 ( 6.8)	0.087
OA_diverse_Citrus = 1 (%)	2 ( 4.3)	1 ( 2.3)	1.000
OA_Artificial = 1 (%)	3 ( 6.5)	4 ( 9.1)	0.951
OA_Scent_Candle_Air_freshener = 1 (%)	2 ( 4.3)	4 ( 9.1)	0.632
OA_Intrusive = 1 (%)	1 ( 2.2)	3 ( 6.8)	0.577
OA_Gum_Toothpaste = 1 (%)	16 ( 34.8)	2 ( 4.5)	0.001
OA_Laundry = 1 (%)	2 ( 4.3)	2 ( 4.5)	1.000
OA_Sauna = 1 (%)	1 ( 2.2)	0 ( 0.0)	1.000
OA_Christmas_baking = 0 (%)	46 (100.0)	44 (100.0)	NA
OA_Clean_Sterile = 1 (%)	2 ( 4.3)	2 ( 4.5)	1.000
OA_sweet = 1 (%)	2 ( 4.3)	16 ( 36.4)	<0.001
OA_Caramel = 1 (%)	0 ( 0.0)	2 ( 4.5)	0.455
OA_refreshing = 1 (%)	7 ( 15.2)	0 ( 0.0)	0.021
OA_Relaxing = 1 (%)	1 ( 2.2)	1 ( 2.3)	1.000

p-values Fisher’s exact for binary oppositions (Vanilla vs. Peppermint)

---

18.3 Visualizing Scent Profiles

---

19 Kimchi and Palmer tasks

Kimchi, R., & Palmer, S. E. (1982). Form and texture in hierarchically constructed patterns. Journal of Experimental Psychology: Human Perception and Performance, 8(4), 521–535. https://doi.org/10.1037/0096-1523.8.4.521

Higher values = global, relational, lower values = local, analytical

(#tab:unnamed-chunk-123) Descriptives on Kimchi-Palmer Mean

ScentNumber	Mean	SD	Min	Max	sample
peppermint	55.03	29.11	0.00	100.00	59
Vanilla	63.62	28.12	0.00	99.92	65

ANOVA Analysis

(#tab:unnamed-chunk-124) ANOVA of Kimchi-Palmer means.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	2.19	1	95	818.61	.143	.022

Note. Note that tne IV here are the scent groups.

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Kimchi_Palmer_total by ScentNumber
## W = 972, p-value = 0.142
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -20.083279   2.666722
## sample estimates:
## difference in location 
##              -8.565718

---

19.1 Dichotomized version

Ijzerman, H., & Semin, G. R. (2009). The thermometer of social relations: Mapping social proximity on temperature. Psychological Science, 20(10), 1214–1220. https://doi.org/10.1111/j.1467-9280.2009.02434.x

(#tab:unnamed-chunk-126) Descriptives on Kimchi-Palmer Mean

ScentNumber	Mean	SD	Min	Max	sample
peppermint	6.77	4.24	0.00	12.00	59
Vanilla	8.18	4.12	0.00	12.00	65

ANOVA Analysis

(#tab:unnamed-chunk-127) ANOVA of Kimchi-Palmer means.

Effect	\(F\)	\(\mathit{df}_1\)	\(\mathit{df}_2\)	\(\mathit{MSE}\)	\(p\)	\(\hat{\eta}^2_G\)
ScentNumber	2.77	1	95	17.47	.099	.028

Note. Note that tne IV here are the scent groups.

## [1] "t-test one-sided p:"

## [1] 0.04972162

Mann-Whitney-U-Test

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Kimchi_Palmer_total_dich by ScentNumber
## W = 935, p-value = 0.08001
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
##  -3.000003e+00  6.391936e-05
## sample estimates:
## difference in location 
##               -1.00001

---

---

20 Mediation models for the assumed psychological mechanisms

20.1 Warm (vs. cool) scent -> warm (vs. cool) scent perceptions -> warm (vs. cool) temperature feelings

## 
## Mediation/Moderation Analysis 
## Call: mediate(y = Temperature_r1 ~ ScentNumber + (CoolWarmRating_1), 
##     data = temp, n.iter = 5000, std = TRUE)
## 
## The DV (Y) was  Temperature_r1 . The IV (X) was  ScentNumber . The mediating variable(s) =  CoolWarmRating_1 .
## 
## Total effect(c) of  ScentNumber  on  Temperature_r1  =  0.28   S.E. =  0.09  t  =  3.22  df=  123   with p =  0.0017
## Direct effect (c') of  ScentNumber  on  Temperature_r1  removing  CoolWarmRating_1  =  0.06   S.E. =  0.08  t  =  0  df=  121   with p =  1
## Indirect effect (ab) of  ScentNumber  on  Temperature_r1  through  CoolWarmRating_1   =  0.22 
## Mean bootstrapped indirect effect =  0.22  with standard error =  0.08  Lower CI =  0.08    Upper CI =  0.39
## R = 0.43 R2 = 0.19   F = 14.1 on 2 and 121 DF   p-value:  6.08e-08 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

## Call: mediate(y = Temperature_r1 ~ ScentNumber + (CoolWarmRating_1), 
##     data = temp, n.iter = 5000, std = TRUE)
## 
## Direct effect estimates (traditional regression)    (c') 
##                  Temperature_r1   se    t  df     Prob
## Intercept                  0.00 0.08 0.00 121 1.00e+00
## ScentNumber                0.06 0.10 0.59 121 5.57e-01
## CoolWarmRating_1           0.40 0.10 4.08 121 8.21e-05
## 
## R = 0.43 R2 = 0.19   F = 14.1 on 2 and 121 DF   p-value:  3.12e-06 
## 
##  Total effect estimates (c) 
##             Temperature_r1   se    t  df    Prob
## ScentNumber           0.28 0.09 3.22 123 0.00165
## 
##  'a'  effect estimates 
##             CoolWarmRating_1   se    t  df     Prob
## Intercept               0.00 0.08 0.00 122 1.00e+00
## ScentNumber             0.55 0.08 7.31 122 3.09e-11
## 
##  'b'  effect estimates 
##                  Temperature_r1  se    t  df     Prob
## CoolWarmRating_1            0.4 0.1 4.09 122 7.67e-05
## 
##  'ab'  effect estimates (through mediators)
##             Temperature_r1 boot   sd lower upper
## ScentNumber           0.22 0.22 0.08  0.08  0.39

## lavaan 0.6-5 ended normally after 21 iterations
## 
##   Estimator                                         ML
##   Optimization method                           NLMINB
##   Number of free parameters                          5
##                                                       
##                                                   Used       Total
##   Number of observations                            90         124
##                                                                   
## Model Test User Model:
##                                                       
##   Test statistic                                 0.000
##   Degrees of freedom                                 0
## 
## Parameter Estimates:
## 
##   Information                                 Expected
##   Information saturated (h1) model          Structured
##   Standard errors                             Standard
## 
## Regressions:
##                      Estimate  Std.Err  z-value  P(>|z|)
##   Temperature_r1 ~                                      
##     ScentNumbr (c)      0.294    0.315    0.934    0.350
##   CoolWarmRating_1 ~                                    
##     ScentNumbr (a)      1.960    0.313    6.255    0.000
##   Temperature_r1 ~                                      
##     ClWrmRtn_1 (b)      0.285    0.088    3.229    0.001
## 
## Variances:
##                    Estimate  Std.Err  z-value  P(>|z|)
##    .Temperature_r1    1.553    0.232    6.708    0.000
##    .CoolWarmRtng_1    2.209    0.329    6.708    0.000
## 
## Defined Parameters:
##                    Estimate  Std.Err  z-value  P(>|z|)
##     ab                0.560    0.195    2.869    0.004
##     total             0.854    0.278    3.075    0.002

20.2 Warm (vs. cool) scent -> warm (vs. cool) scent perceptions -> warm (vs. cool) temperature feelings -> preference for luxury coffee

## 
## Mediation/Moderation Analysis 
## Call: mediate(y = CoffeePreference_r1 ~ ScentNumber + (CoolWarmRating_1) + 
##     (Temperature_r1), data = temp, n.iter = 5000, std = TRUE)
## 
## The DV (Y) was  CoffeePreference_r1 . The IV (X) was  ScentNumber . The mediating variable(s) =  CoolWarmRating_1 Temperature_r1 .
## 
## Total effect(c) of  ScentNumber  on  CoffeePreference_r1  =  0.21   S.E. =  0.09  t  =  2.4  df=  123   with p =  0.018
## Direct effect (c') of  ScentNumber  on  CoffeePreference_r1  removing  CoolWarmRating_1 Temperature_r1  =  0.18   S.E. =  0.09  t  =  0  df=  120   with p =  1
## Indirect effect (ab) of  ScentNumber  on  CoffeePreference_r1  through  CoolWarmRating_1 Temperature_r1   =  0.03 
## Mean bootstrapped indirect effect =  0.03  with standard error =  0.09  Lower CI =  -0.12    Upper CI =  0.21
## R = 0.25 R2 = 0.06   F = 2.77 on 3 and 120 DF   p-value:  0.0304 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

## Call: mediate(y = CoffeePreference_r1 ~ ScentNumber + (CoolWarmRating_1) + 
##     (Temperature_r1), data = temp, n.iter = 5000, std = TRUE)
## 
## Direct effect estimates (traditional regression)    (c') 
##                  CoffeePreference_r1   se     t  df   Prob
## Intercept                       0.00 0.09  0.00 120 1.0000
## ScentNumber                     0.18 0.11  1.71 120 0.0892
## CoolWarmRating_1               -0.02 0.11 -0.22 120 0.8300
## Temperature_r1                  0.15 0.10  1.57 120 0.1190
## 
## R = 0.25 R2 = 0.06   F = 2.77 on 3 and 120 DF   p-value:  0.0448 
## 
##  Total effect estimates (c) 
##             CoffeePreference_r1   se   t  df   Prob
## ScentNumber                0.21 0.09 2.4 123 0.0181
## 
##  'a'  effect estimates 
##             CoolWarmRating_1   se    t  df     Prob
## Intercept               0.00 0.08 0.00 122 1.00e+00
## ScentNumber             0.55 0.08 7.31 122 3.09e-11
##             Temperature_r1   se   t  df    Prob
## Intercept             0.00 0.09 0.0 122 1.00000
## ScentNumber           0.28 0.09 3.2 122 0.00172
## 
##  'b'  effect estimates 
##                  CoffeePreference_r1   se     t  df  Prob
## CoolWarmRating_1               -0.02 0.11 -0.22 121 0.829
## Temperature_r1                  0.15 0.10  1.58 121 0.118
## 
##  'ab'  effect estimates (through mediators)
##             CoffeePreference_r1 boot   sd lower upper
## ScentNumber                0.03 0.03 0.09 -0.12  0.21
## 
##  'ab' effects estimates for each mediator for CoffeePreference_r1 
##             CoolWarmRating_1   sd lower upper
## ScentNumber            -0.01 0.04  -0.1  0.09
##             Temperature_r1   sd lower upper
## ScentNumber           0.04 0.04 -0.03  0.14

20.3 Warm (vs. cool) scent -> relational processing (Kimchi-Palmer) -> preference for luxury coffee

## 
## Mediation/Moderation Analysis 
## Call: mediate(y = CoffeePreference_r1 ~ ScentNumber + (Kimchi_Palmer_total_dich), 
##     data = temp, n.iter = 5000, std = TRUE)
## 
## The DV (Y) was  CoffeePreference_r1 . The IV (X) was  ScentNumber . The mediating variable(s) =  Kimchi_Palmer_total_dich .
## 
## Total effect(c) of  ScentNumber  on  CoffeePreference_r1  =  0.21   S.E. =  0.09  t  =  2.4  df=  123   with p =  0.018
## Direct effect (c') of  ScentNumber  on  CoffeePreference_r1  removing  Kimchi_Palmer_total_dich  =  0.19   S.E. =  0.09  t  =  0  df=  121   with p =  1
## Indirect effect (ab) of  ScentNumber  on  CoffeePreference_r1  through  Kimchi_Palmer_total_dich   =  0.02 
## Mean bootstrapped indirect effect =  0.02  with standard error =  0.03  Lower CI =  -0.03    Upper CI =  0.09
## R = 0.24 R2 = 0.06   F = 3.61 on 2 and 121 DF   p-value:  0.0154 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

## Call: mediate(y = CoffeePreference_r1 ~ ScentNumber + (Kimchi_Palmer_total_dich), 
##     data = temp, n.iter = 5000, std = TRUE)
## 
## Direct effect estimates (traditional regression)    (c') 
##                          CoffeePreference_r1   se    t  df   Prob
## Intercept                               0.00 0.09 0.00 121 1.0000
## ScentNumber                             0.19 0.09 2.15 121 0.0336
## Kimchi_Palmer_total_dich                0.11 0.09 1.22 121 0.2230
## 
## R = 0.24 R2 = 0.06   F = 3.61 on 2 and 121 DF   p-value:  0.0301 
## 
##  Total effect estimates (c) 
##             CoffeePreference_r1   se   t  df   Prob
## ScentNumber                0.21 0.09 2.4 123 0.0181
## 
##  'a'  effect estimates 
##             Kimchi_Palmer_total_dich   se    t  df   Prob
## Intercept                       0.00 0.09 0.00 122 1.0000
## ScentNumber                     0.17 0.09 1.89 122 0.0611
## 
##  'b'  effect estimates 
##                          CoffeePreference_r1   se    t  df  Prob
## Kimchi_Palmer_total_dich                0.11 0.09 1.23 122 0.221
## 
##  'ab'  effect estimates (through mediators)
##             CoffeePreference_r1 boot   sd lower upper
## ScentNumber                0.02 0.02 0.03 -0.03  0.09