What Percentage Of Women Use Makeup Everyday In America Academic Papers

• Journal List
• PLoS One
• PMC5058481

PLoS One. 2016; 11(ten): e0164218.

Facial Cosmetics and Attractiveness: Comparing the Outcome Sizes of Professionally-Applied Cosmetics and Identity

Alex L. Jones

¹ Department of Psychology, Swansea University, Swansea, United Kingdom

Robin S. S. Kramer

ⁱⁱ Section of Psychology, University of York, York, United Kingdom

Katsumi Watanabe, Editor

Received 2016 Apr 19; Accepted 2016 Sep 21.

Supplementary Materials

S1 Dataset: Information from Study 1. Each participant rated all 33 YouTube models, but each model appeared in a randomly selected cosmetics condition. All weather are stated in the information. We averaged across participants for each image, edifice a score for each identity under both cosmetics conditions.

(XLS)

GUID: B371FB9A-96FB-439F-982C-433AD599528E

S2 Dataset: Data from Written report ii. Each participant rated all 45 supermodels, but each model appeared in a randomly selected cosmetics condition. All conditions are stated in the information. We averaged across participants for each image, building a score for each identity under both cosmetics weather.

(XLS)

GUID: B0DB4D61-2965-4F4A-B6B8-0CDDE3B85F98

S3 Dataset: Data from the quantity raters in both studies. Canvas ane contains the quantity information from Written report 1, and Sheet two contains the quantity data for Report 2. Participants compared each model without and with cosmetics, indicating how much cosmetics the faces were wearing.

(XLS)

GUID: E9DEBD5E-9F52-4297-91CB-9406ABC44EED

Data Availability Statement

All relevant data are inside the paper and its Supporting Information files.

Abstract

Forms of body ornament be in all human cultures. However, in Western societies, women are more likely to engage in appearance modification, especially through the use of facial cosmetics. How effective are cosmetics at altering attractiveness? Previous research has hinted that the effect is not large, particularly when compared to the variation in attractiveness observed between individuals due to differences in identity. In lodge to build a fuller understanding of how cosmetics and identity impact bewitchery, hither we examine how professionally-practical cosmetics change attractiveness and compare this result with the variation in bewitchery observed between individuals. In Study 1, 33 YouTube models were rated for attractiveness before and afterward the application of professionally-applied cosmetics. Cosmetics explained a larger proportion of the variation in attractiveness compared with previous studies, but this event remained smaller than variation acquired by differences in bewitchery between individuals. Study ii replicated the results of the starting time study with a sample of 45 supermodels, with the aim of examining the effect of cosmetics in a sample of faces with low variation in attractiveness between individuals. While the effect size of cosmetics was more often than not large, betwixt-person variability due to identity remained larger. Both studies also institute interactions between cosmetics and identity–more than attractive models received smaller increases when cosmetics were worn. Overall, nosotros show that professionally-applied cosmetics produce a larger result than cocky-practical cosmetics, an of import theoretical consideration for the field. However, the effect of individual differences in facial advent is ultimately more than of import in perceptions of attractiveness.

Introduction

Modification of the torso with dyes, paints, and other pigments is among the almost universal of human being behaviours, present in all cultures [1–3]. Nevertheless, in Western society, women perform the majority of self-adornment [four], and perhaps the nigh prevalent behaviour of this kind is the use of facial cosmetics. This behaviour is served by the global cosmetics industry which is worth billions of pounds [5].

Women report using cosmetics for a variety of reasons, ranging from anxiety about facial appearance, conformity to social norms, and public self-consciousness [half dozen–8], through to appearing more sociable and assertive to others [half-dozen]. Cosmetics are constructive at improving social perceptions that the wearer may wish to attune, with individuals appearing to be healthier and earning more [9], displaying greater competence, likeability and trustworthiness [10], likewise as appearing more than prestigious and dominant [xi]. Cosmetics likewise influence the behaviour of others, especially men, who tip higher amounts and with greater frequency to waitresses wearing cosmetics [12], and are more than likely to approach wearers in the environment [13]. It is likely that the event of cosmetics on social perceptions is brought well-nigh by the increment in attractiveness information technology confers to faces, which is now a well documented effect [ten,xiv–17]. Research has documented cosmetics function by altering sex-typical colouration in faces such equally facial dissimilarity [18–21], by increasing the homogeneity of facial pare [22,23], or by affecting colour cues to traits such equally health [24] and age [25].

While the outcome of cosmetics on perceived attractiveness seems clear [14,17], other research has revealed it is more nuanced than previously thought. Etcoff and colleagues [ten] demonstrated that attractiveness increased linearly with the amount of cosmetics worn—but, more than cosmetics equates to actualization more than attractive. Of the range of cosmetics that tin be worn, the quantity of cosmetics applied to the eyes and mouth accept been shown to exist pregnant predictors of attractiveness [26], with more than cosmetics on these features leading to higher ratings of bewitchery. However, other evidence suggests that the typical amount of cosmetics applied by a sample of young women is excessive, with observers preferring close to half the actual corporeality for optimal attractiveness [xvi], calling into question the linear human relationship between cosmetics quantity and attractiveness.

One concern of facial attractiveness research is that it does not compare the effects of predictors of attractiveness (e.chiliad., symmetry, averageness, sex typicality [27]; against other sources of variation [28]. Recent work has begun to address this by examining the importance of within-person variation in bewitchery (acquired by the presence or absence of makeup, for example), compared with the between-person variation in attractiveness simply due to differences between identities [29]. Specifically, it has been previously shown that the issue of cosmetics on attractiveness, a source of within-person variation, is very small, explaining merely 2% of the variance in ratings [15]. This is an especially minor effect when compared with differences in bewitchery between individuals, a between-person variation in attractiveness, which explained 69% of the variance in judgements. More but, while facial cosmetics practice increment attractiveness, that contribution is pocket-sized and does little to modify an individual'southward attractiveness standing in the population.

Withal, the use of cosmetics is an idiosyncratic and extremely varied practice [3], and its effect on attractiveness is more complex than previously thought. The apply of a professional makeup artist is a common practice in almost all studies examining the effect of cosmetics on perceptions [9,ten,12,17,thirty,31], and only a few utilise cocky-applied cosmetics [14,16,26]. An initial examination of the event size of cosmetics on attractiveness also had models self-employ their cosmetics [15]. There are good reasons for using professionally-applied cosmetics, as it provides a clearer test of how cosmetics alter facial attractiveness. The increased variability in self-applied cosmetics, due, for example, to differences in application skill or the products used, could make information technology more difficult to find an effect of cosmetics on attractiveness, and previous work has indeed found the consequence to exist small [15]. This distinction represents a trade-off between experimental control and ecological validity—the vast majority of women, if whatsoever, do not have a professional person makeup artist apply their cosmetics daily, however the majority of studies examining cosmetics and attractiveness draw conclusions based on professionally-applied cosmetics, which may just indirectly inform as to how cosmetics affect attractiveness in the real earth.

We seek to address important theoretical points regarding how cosmetics influence attractiveness. How large is the event size of cosmetics on attractiveness when cosmetics take been professionally-practical? If cosmetics in psychological experiments are applied with more skill than is typically accomplished, then electric current knowledge of cosmetics and attractiveness likely overstates the relationship, given the reliance on professionally-applied cosmetics in the literature. Moreover, how does the ability of professionally-applied cosmetics compare to previous measures of the effect of cosmetics on attractiveness? In the following study, nosotros examine the effect size of cosmetics on attractiveness in two sets of faces that accept had cosmetics applied professionally, with the prediction that the upshot will be substantially larger than the previous assessment that considered cocky-applied cosmetics [15]. In addition, by using a similar pattern to previous research, nosotros can depict direct comparisons with current knowledge of how cosmetics and identity affect attractiveness.

A carve up but related question regarding cosmetics concerns how information technology affects faces of different levels of bewitchery. Many studies in the literature on cosmetics and social perceptions have used models recruited from academy or higher [14,15,20]. How practice cosmetics bear upon faces of a unlike population, specifically faces considered to exist very attractive? Previous research found no interaction between cosmetics and identity [15], suggesting cosmetics bear on each face's attractiveness similarly. Nonetheless, the models used were of a university-aged sample of population-typical attractiveness levels. The present studies, particularly Report 2, examine the effect cosmetics take on perceived attractiveness in a sample of women typically considered to be very attractive—models. Using a sample of faces that are already constrained in attractiveness enables us to manipulate another source of variation in attractiveness, specifically between-person variability. As such, nosotros tin observe the effects of cosmetics on attractiveness in a sample with a (hypothesised) lower effect of identity (differences between individuals) than elsewhere.

The present study has several aims. Commencement, we examine how cosmetics bear on attractiveness when cosmetics have been professionally-applied. We predict that cosmetics will have a notably larger effect size in this sample compared to the previous report examining this question [fifteen]. Second, we consider the effect size of cosmetics in sets of faces that are considered highly bonny, where between-person variation (identity effect size) should be reduced. The relative effect size of cosmetics may therefore be increased, and may be more than likely to overshadow the smaller between-person variation in bewitchery. Conversely, cosmetics may accept less of an effect in these samples as the women are already at the higher end of attractiveness without cosmetics, leaving little room for judgements of bewitchery to increase when cosmetics are applied. Finally, past using an identical pattern to previous enquiry [15], nosotros will compare the findings obtained in these studies to those presented in previous research in order to build a fuller motion-picture show of the relative importance of cosmetics and identity in attractiveness perceptions.

Study i

In the offset written report, nosotros examine how cosmetics impact attractiveness when they are applied professionally. To do this, we take reward of an Net-based sample to acquire images of models whose cosmetics have been applied past loftier-profile makeup artists. Compared to previous work examining this question [15], we predict that the issue size due to cosmetics should be larger here. However, the consequence size of identity may still overshadow it.

Method

Participants

Xc Northward American academy students (historic period M = xviii.57 years, SD = 0.75, 41 men) participated in the main study for course credit. Due to a software error, age data was not recorded for the beginning 50 participants, with the mean age being calculated from the remaining participants. However, all participants were within the same demographic and historic period range. A farther fifteen students (age M = 19.93 years, SD = i.sixteen, three men) rated the quantity of cosmetics worn by the models. Informed consent was obtained from all participants included in the study.

Ideals Statement

Ethical approval for all studies was obtained from the Gettysburg College institutional review board (IRB). All participants gave written informed consent before beginning the report.

Stimuli

From the YouTube website, we collected images of White British women (n = 33, age unknown merely approximately 20–35 years), who acted as models while their cosmetics were practical by loftier-profile professional person makeup artists from the United Kingdom. Xx-three models were obtained from one creative person's aqueduct (world wide web.youtube.com/user/lisaeldridgedotcom) with a further ten collected from another (www.youtube.com/user/ctilburymakeup). Nosotros utilised all available videos at the time of writing that featured a model receiving a makeover where they were shown earlier and after an application of cosmetics. In addition, we included simply videos where faces began costless of cosmetics, and the artist had the intention of applying a particular cosmetics look, rather than with the aim of hiding blemishes or pare weather (such equally acne). Images were captured from video tutorials, which served to instruct viewers on a number of pop cosmetics styles for a range of scenarios. Both authors classified the cosmetics looks into categories using data provided past descriptions inside the videos. Iii categories were credible—an everyday, natural look (northward = 7), a 'going out' wait (n = 14), and vintage or editorial looks based on cosmetics the makeup artist had applied during professional person photo shoots in the by (n = 12). A third researcher, with extensive experience in this field, arrived at these three categories independently, providing farther confirmation.

We captured a high-resolution screenshot of each model at the finish of each video, where images of the models were presented before and after their application of cosmetics side-past-side. Models had a neutral expression and looked straight into the camera for the comparison. In addition, the 2 photographs were taken nether the same lighting and photographic camera conditions. From each comparing screenshot, we cropped the 'before' and 'subsequently' versions of each model to produce two separate images. Final images were cropped but below the mentum, at the hairline (or mid-forehead based on the limitations of the original), and tight to the widest part of the confront (and and then removing the ears). Given the variable nature of the images in terms of hairstyle, we chose models whose hair did not occlude their faces, and nosotros masked loose pilus in the lower portions of the images if it was not tied back. Images were resized to a peak of 451 pixels. Given copyright restrictions, we present the average of models without cosmetics, and separately with cosmetics, in Fig 1 to illustrate. Averages were produced using JPsychomorph after landmarks were applied to the facial features in each image [32].

The average model without (left) and with cosmetics (right).

These averages are cropped mid-brow because several of the YouTube videos presented individuals in this way, resulting in insufficient information above this point for generating averages.

Procedure

Participants rated the attractiveness of the models using custom PsychoPy software [33]. Images were presented in a random order, and each participant rated each model just in one case, in a randomly selected cosmetics condition (i.due east., either with or without cosmetics). This design was specifically chosen to prevent carryover effects between conditions [fifteen,29]. Participants rated the attractiveness of the models on a 1 (very unattractive) to 7 (very attractive) scale, indicating their response via mouse click. Stimuli remained onscreen until a judgement was made.

A separate sample of participants judged the quantity of cosmetics worn by the models. These participants saw the 'without' and 'with cosmetics' images onscreen next to each other, and were asked 'how much makeup has been applied to this face?' Participants indicated their responses via mouse click on a one (very light) to 7 (very heavy) calibration. Trials were presented in a random order. Though this is only a perceived mensurate of quantity, rather than an actual quantity of cosmetics, nosotros believe it to exist suitable as it is the perceived quantity that would affect the perceptions of observers. Chiefly, other studies have found general agreement in the quantity of cosmetics practical by a professional person makeup artist and the perceived corporeality of cosmetics beingness worn [31].

Results

Each prototype was rated an average of 45 times (SD = 4.45). Nosotros examined agreement by calculating the pooled standard deviation for ratings in each cosmetics status; without SD _p = 1.34; with cosmetics SD _p = 1.44. Responses were given on a 7-point calibration, so the generally low variability indicates practiced agreement in ratings [15,34]. To examine effects of observer sex on ratings, the data were carve up by the sexual practice of each observer earlier averaging. This resulted in four scores for each model—one in each cosmetics condition, as rated past men and women.

Nosotros as well calculated the average amount of perceived cosmetics applied (M = 4.96, SD = 1.09), equally judged by the split up sample of raters. These judgements of quantity were collected in order to be able to control for the varying amounts of cosmetics worn by each model in our analyses. All the same, this measure showed no relationship with the dependent variable (attractiveness) at all levels of observer sexual practice and cosmetics, all rs < .25, ps > .160. Equally such, there was no reason to include quantity every bit a covariate, and nosotros therefore analysed our results using a repeated measures ANOVA with model as the unit of measurement of analysis.

We focus here on the effect sizes of variables in order to estimate the existent world effect of cosmetics on bewitchery. In particular, we utilize eta squared (η ² ) equally a measure of upshot size, which expresses how much each factor contributes to the total variance in bewitchery ratings equally an interpretable percentage value, rather than partial eta squared, which does not sum beyond factors to one. We calculated η ² effect sizes for both main effects (Cosmetics, Observer Sex) and the interaction past dividing the sums of squares (SS) attributable to each effect by the total SS, calculated by summing the SS attributable to each effect and their respective errors. Nosotros also gave special consideration to the variance attributable to differences between items. This variation is typically ignored in repeated measures analyses since it normally represents variation between participants on the measured dependent variable, which is generally unimportant for repeated measures designs (which instead focus on variation within participants). However, in this case, it takes on a useful property. By using the images of the models as the unit of assay, the variation between models represents variation in attractiveness arising due to the fact that models accept different facial identities or appearances. Nosotros were therefore able to calculate an effect size for this 'identity' measure. The full results of the ANOVA are reported in Table 1, illustrating the effect sizes, their associated SS, and other statistics. It should be noted that there is no fault term for conducting an F test on differences betwixt models, and as such, no F ratio is calculated interactions with the Identity measure out can exist interpreted every bit an error term for that variable [35].

Table 1

Results of the analysis of variance from Study 1.

Source	df	SS	η ii	F	p
Identity (I)	32	61.27	0.45
Observer Sexual practice	1	0.85	0.01	10.03	.003
Observer Sex × I	32	2.70	0.02
Cosmetics	1	44.83	0.33	76.33	< .001
Cosmetics × I	32	18.79	0.fourteen
Observer Sex × Cosmetics	ane	one.29	0.01	8.17	.007
Observer Sex × Cosmetics × I	32	5.05	0.04
Total	131	134.78

Men assigned lower ratings of attractiveness (Grand = iii.74, 95% CI [3.47, 4.00]) than women (M = three.89, [3.66, 4.13]), a result consequent with previous literature [15,36,37] which we exercise not pursue farther here. Chiefly, models were rated equally more attractive with cosmetics (Grand = iv.39, [4.eleven, four.68]) than without (M = 3.23, [2.95, 3.51]). The Observer Sex x Cosmetics interaction was driven by men rating faces without cosmetics every bit less bonny than women rating those same faces, t(32) = iv.32, p < .001, d = 0.75, but both sexes assigned similar ratings for models with cosmetics, t(32) = 0.42, p = .676, d = 0.07, indicating a larger influence of cosmetics on bewitchery for men. However, the event size of this interaction was very small-scale (η ⁱⁱ = 0.01), suggesting a relatively unimportant issue.

Of more importance was the Cosmetics x Identity interaction (η ² = 0.fourteen), which indicates that the application of cosmetics altered the attractiveness of private models differently. To examine this further, we computed a deviation score for each model between their bewitchery with and without cosmetics, every bit rated by men and women. This difference illustrates the boost in attractiveness conferred by cosmetics, and we carried out a correlation between these values and the attractiveness of the models without cosmetics. Ratings assigned by both women and men showed a negative correlation betwixt these values, r(31) = -.53, 95% CI [-.73, -.23], p = .001, and r(31) = -.48, [-.71, -.16], p = .005, respectively, indicating that the more attractive a model was, the less of an increase in bewitchery cosmetics conferred, a pattern which did non change when combining ratings given past men and women, r(31) = -.46, [-.69, -.xiv], p = .007 (see Fig two).

An analogy of the average attractiveness (combining ratings fabricated by men and women) of each model, both without cosmetics and with cosmetics.

Models are ordered in terms of increasing attractiveness without cosmetics. An upward pointing pointer indicates an increase in attractiveness with cosmetics, while a downwards pointer indicates a subtract.

Tabular array i illustrates that the Identity result size (η ² = 0.45) is 1.36 times larger than the effect size attributed to Cosmetics (η ⁱⁱ = 0.33). The differences in attractiveness between individuals explains more variance than an awarding of cosmetics, but the ratio of these effect sizes is much smaller than in previous accounts [fifteen]. This suggests that a professional application of cosmetics (in comparison with cocky-application) is capable of producing a larger effect on attractiveness perceptions, although this remains smaller than the effect due to identity differences betwixt women.

We conducted a concluding analysis to examine whether the cosmetics 'look' ascribed by the artist affected perceptions of attractiveness differently for men and women. The above analysis was repeated, but with the improver of 'await' equally a source of variation between models. The three-fashion mixed model ANOVA revealed no significant primary effects of cosmetics look or interactions with this gene, all Fs < one.18, psouthward > .320. However, it is worth noting that the 'cosmetics wait' variable had low power (ranging from .076 to .242 across chief effects and interactions), then farther study is required to investigate the office of cosmetics expect in perceived attractiveness.

Report 2

The models used in Study one were women who had agreed to participate for the purposes of sit-in in a makeup tutorial. We have shown that the outcome of cosmetics, when professionally-practical, results in a larger effect size compared with previous research [fifteen]. Next, we investigate how cosmetics alter the attractiveness of a sample of women who are generally regarded every bit very attractive and earn a living based on their appearance—supermodels. We examine how much variation in attractiveness can exist explained by cosmetics, and compare it with the outcome size of identity, the differences in bewitchery betwixt supermodels. Here, the issue size of identity should be smaller, given the potentially homogenous nature of the women in terms of attractiveness. How much of a do good do cosmetics confer to highly attractive women, and in turn, do cosmetics overcome the differences in attractiveness between individuals?

Method

Participants

One hundred new participants completed the study for course credit (age 1000 = 19.28 years, SD = i.46, 46 men), fourteen of which were students at a Scottish academy (age Chiliad = 19.28 years, SD = 1.68, one man), while the rest were students at a Due north American university (historic period K = 19.28 years, SD = 1.05, 45 men). A further sample of 14 N American students from the same university (age M = twenty.50 years, SD = 1.28, ii men) rated the quantity of cosmetics worn by the models.

The removal of the fourteen participants from the Scottish university (who alive in the Uk rather than the U.s.) did not alter the pattern of results described below, aside from producing a meaning main effect of Observer Sex, F(one, 44) = eighteen.64, p < .001, η ² = .02. As in Study 1, men provided lower ratings of attractiveness (1000 = iv.07, [3.91, iv.23]) than women (One thousand = 4.32, [iv.12, four.53]). Yet, as this is a well-demonstrated effect and did not alter the presence of the interaction between cosmetics and observer sexual activity, nosotros include these extra participants for the additional validity they confer.

Ideals Statement

Ethical approving for all studies was obtained from the Gettysburg College institutional review board (IRB). All participants gave written informed consent before beginning the report. The Ethical Governance and Approving Organisation at the University of Aberdeen granted approving for the study conducted there. Again, all participants gave written informed consent before first the study.

Stimuli

We collected images (n = 45) of supermodels without their makeup from the Internet. These images were casting photographs for Louis Vuitton's Fall-Wintertime 2010 runway show. All pictures were taken with the models looking directly into the camera, with a neutral expression. We then collected images of the same women wearing cosmetics from professional photograph shoots, and selected images where they had a neutral expression and were looking directly into the photographic camera in gild to lucifer the casting photographs as closely as possible. However, these cosmetics photos were considerably less constrained in that the lighting varied betwixt images, as did the corporeality of time betwixt the two photos for each model. Therefore, while every care was taken to ensure similarity between these images and those of Study one, we annotation that such limitations mean that any conclusions fatigued from this report are necessarily more tentative.

Concluding images were cropped as in Study 1 to just below the chin, at the hairline, and tight to the widest part of the face (and and then removing the ears). Hair was masked at the bottom of the images every bit before, and images were resized to a height of 250 pixels. Given copyright restrictions, nosotros present the average of supermodels without cosmetics, and separately with cosmetics, in Fig 3 to illustrate.

The average supermodel without (left) and with cosmetics (correct).

Procedure

Nosotros used the aforementioned procedure as in Report 1. However, given that the photographs were of supermodels, there was a adventure they would exist recognised by observers. As such, we added a 'recognise' option onscreen where participants could point their recognition of the model rather than providing a rating of attractiveness. Familiarity with the models may outcome in unwanted influences on ratings. Beyond all images, an average of four.95 trials were skipped (SD = 2.97). Ratings of attractiveness were therefore just nerveless for models that were not recognised by the raters.

Results

Each image received an average of 50 ratings (SD = 4.68). Agreement was calculated equally earlier, using a pooled standard deviation for ratings within each cosmetics condition, and showed generally higher levels (lower variation) than Study one; without cosmetics SD _p = 1.xi; with cosmetics SD _p = 1.32. We split the data by the sexual practice of the observer as before, and computed the average rating for each model in both cosmetics conditions equally assigned by men and women.

We then averaged the ratings of quantity assigned by the carve up sample of raters (Thou = iv.29, SD = 1.17) for use equally a covariate in subsequent analyses. However, as in Study ane, the quantity measure showed no relationship with the dependent variable at any levels of each independent variable, all rs < .11, pdue south > .476. Equally such, analyses were carried out without inclusion of this covariate using repeated measures ANOVA, the results of which are summarised in Table 2. We too compared the perceived quantity ratings of the faces in Study 1 to the faces here, finding that the sample of supermodels (Chiliad = 4.29, [3.95, 4.63]) were perceived every bit wearing less cosmetics than the YouTube models (Yard = 4.96, [4.57, 5.36]), t(76) = 2.56, p = .012, d = 0.59.

Table 2

Results of the analysis of variance from Study 2.

Source	df	SS	η 2	F	p
Identity (I)	44	58.08	0.43
Observer Sex	1	0.37	0.00	ii.79	.102
Observer Sex × I	44	5.85	0.04
Cosmetics	1	33.35	0.25	47.89	< .001
Cosmetics × I	44	30.64	0.23
Observer Sex × Cosmetics	1	ane.73	0.01	20.66	< .001
Observer Sex × Cosmetics × I	44	3.68	0.03
Total	179	133.seventy

As before, models were rated as more attractive with cosmetics (M = iv.53, [4.28, 4.77]) than without (K = iii.67, [3.49, 3.85]). The Observer Sex x Cosmetics interaction was over again driven by men rating faces without cosmetics equally less attractive than women rating those aforementioned faces, t(44) = 4.65, p < .001, d = 0.69, with both sexes perceiving the models as similarly attractive with cosmetics, t(44) = 1.37, p = .176, d = 0.21. All the same, as before, the effect size of this interaction was small (η ² = 0.01).

The upshot size of the Cosmetics 10 Identity interaction (η ² = 0.23) was almost as large as the effect of cosmetics itself (η ^two = 0.25), indicating the awarding of cosmetics affected the attractiveness of the supermodels differently. As earlier, we computed a divergence score (for men and women's ratings separately) betwixt cosmetics weather, and correlated this score with the attractiveness of the supermodels without cosmetics. Again, there was a negative correlation between the boost in bewitchery with cosmetics and the bewitchery of the models without cosmetics, for both women r(43) = -.40, [-.62, -.12], p = .006, and men r(43) = -.42, [-.64, -.xiv], p = .004, as well as when ratings given past both sexes were combined, r(43) = -.40, [-.62, -.12], p = .004. As before, this indicates that the more attractive the supermodel is perceived to be, the less of a boost in attractiveness cosmetics confer. This is illustrated in Fig four.

An illustration of the average attractiveness (combining ratings made by men and women) of each model, both without cosmetics and with cosmetics.

Models are ordered in terms of increasing attractiveness without cosmetics. An up pointing arrow indicates an increment in bewitchery with cosmetics, while a downward arrow indicates a subtract.

The effect size of Identity (η ⁱⁱ = 0.43), due to differences in attractiveness betwixt supermodels, was 1.72 times greater than that of cosmetics (η ^two = 0.25), a ratio slightly larger than that observed in Report 1. Differences in attractiveness between individuals explained more variance than cosmetics, even among a sample of women regarded equally highly attractive. The issue size of Cosmetics in this study was smaller than that found in Report 1, suggesting that cosmetics may produce a smaller increase in bewitchery for women who are already at the top finish of the bewitchery scale, although the mean ratings for faces exercise non propose a ceiling effect.

It is also worth noting that the effect size of Identity in this study was very shut to the value reported in Study one, which goes against our prediction that a sample of supermodels should have smaller betwixt-person variability in attractiveness. All the same, this value is however notably smaller than the variation between individuals in a sample of academy students [15], supporting our prediction of reduced between-person variability.

Combined Analyses

We accept shown that professionally-applied cosmetics increment the bewitchery of both models and supermodels, with by and large larger outcome sizes than take been observed elsewhere [15]. Hither, we combine the data from Study 1 with the data reported in previous work that provided an estimate of the outcome size of cosmetics when self-practical to a pupil population [15]. This will allow a comparing of both model sets without and with cosmetics, and an overall comparing of the upshot size of cosmetics and identity in a pooled setting of cosmetics use. We included simply the models from Report i as these images were captured under more than controlled conditions, similar to the images used in the previous work. In the initial study [fifteen], there were 44 self-reported White women acting as models (age One thousand = 21.18, SD = 1.94). Models practical their ain cosmetics from a range of provided products, and were rated using the aforementioned procedure used here. See [15] for total details.

To conduct this analysis, we employed a three-way mixed ANOVA: Fix (Students, YouTube) × Cosmetics (With, Without) × Observer Sex (Female, Male person). Set represented a between-subjects factor, while the remaining factors were both within-subjects. As before, the model was the unit of analysis. Since a factorial ANOVA produces several statistical tests, we focus on the theoretically important outcomes. In this case, an interaction between Gear up and Cosmetics indicates that an application of cosmetics affects the model sets differently. Nosotros would predict models that received an application of professional person cosmetics would appear more bonny.

As observed across the original data [15] and the two studies presented here, at that place was a main event of Observer Sexual activity, F(one, 75) = 122.45, p < .001, η ² = .04, following the usual pattern of men (G = 3.21, [three.05 three.37]) assigning lower ratings than women (1000 = 3.61, [3.44, three.78]). Models as well received higher ratings of attractiveness when viewed with cosmetics (1000 = three.76, [iii.58, iii.95]) compared to when they were viewed without (M = 3.05, [2.88, 3.23]), F(i, 75) = 97.35, p < .001, η ^two = .12. Models from the YouTube set up were besides rated as more attractive (M = three.82, [3.57, 4.06]) than those in the pupil prepare (M = 3.00, [2.79, 3.21]), F(one, 75) = 24.95, p < .001, η ² = .16.

The predicted interaction between Set and Cosmetics was present, F(1, 75) = forty.59, p < .001, η ² = .05. Bonferroni adapted mail-hoc tests revealed that without cosmetics, the YouTube models (M = 3.23, [2.97, three.49]) were rated as slightly more than attractive than models from the student prepare (One thousand = 2.87, [2.65, 3.09]), p = .041, d = 0.24. However, with cosmetics, YouTube models (M = iv.39, [four.12, 4.67]) received significantly higher ratings of attractiveness than the pupil models (M = three.12, [2.88, 3.37]), p < .001, d = 0.79, indicating a larger change in bewitchery with professionally-practical cosmetics than with cocky-applied cosmetics.

We can also depict comparisons between the sizes of our effects across all three studies (the ii presented here and the student set up). While η ² is platonic for comparison effect sizes within a study (the full always sums to 100%), comparison between studies is by and large not recommended because the full variability depends on the written report design and the number of contained variables [38]. All the same, the two studies reported here, as well as earlier information [15], employ identical study designs, and the total variability is very similar in all cases (Report 1 SS full = 134.76, Study 2 SS total = 133.seventy, [15] SS full = 129.23). The main differences were the models used and the type of cosmetics practical. Equally such, we can justifiably make some comparisons between the upshot sizes of cosmetics and identity beyond these studies.

While the effect size due to identity was similar in Studies 1 and 2 (η ⁱⁱ = .45 and η ² = .43, respectively), the earlier study using students showed a much larger result (η ² = .69). The issue size of cosmetics in Studies 1 and 2 (η ² = .33 and η ⁱⁱ = .25, respectively), in contrast, were much larger than in the student study (η ² = .02). Therefore, while variation in attractiveness between individuals was somewhat greater among a sample of university students every bit compared to models and supermodels (as nosotros would expect), the effect size of professionally-applied cosmetics was much larger than self-applied cosmetics. It is also important to note that the consequence sizes obtained for the data in Report two are to be interpreted cautiously, given the more unconstrained nature of the images.

Full general Discussion

Across several studies, nosotros find that using cosmetics increases perceptions of attractiveness compared to no cosmetics, with several novel findings and caveats. Commencement, we bear witness that the upshot size of cosmetics on bewitchery is large when those cosmetics have been professionally-applied, though the upshot of identity is still greater. However, the deviation between identity and cosmetics effects is much smaller than in a student sample of faces with self-applied cosmetics [xv]. 2d, we show that in a sample of supermodels with a smaller, more constrained result size of identity (i.e., reduced betwixt-person variance in attractiveness), identity is still more important than cosmetics, though the event size of cosmetics is all the same larger than in previous cases. In both cases, only particularly the set of supermodels, we institute testify of an interaction betwixt facial identity and cosmetics, indicating a differential outcome of cosmetics on bewitchery. Further analysis revealed that the more than bonny a confront was without cosmetics, the less of an increase in bewitchery cosmetics conferred.

Beyond all studies, nosotros observed that the effect of facial identity was larger than the effect of cosmetics. This finding extends previous enquiry demonstrating that between-person variation is consistently larger than within-person manipulations of attractiveness [15,29]. Interestingly, the ratio between the result sizes of identity and cosmetics in these studies (i.e., how much more variation identity explained than cosmetics in attractiveness judgements) is smaller than the comparison observed with emotional expression [29], suggesting that professionally-a—pplied cosmetics might be more effective at modulating bewitchery perceptions than facial expression, at least in female person faces. Additionally, the finding that identity might be more important than inside-person variation should perhaps exist interpreted with caution. We refer to 'identity' in the current newspaper but use single, passport-mode images of each model. Notwithstanding, individuals appear differently across dissimilar photographs, and this within-person variation in appearance has besides been shown to affect perceived attractiveness [39].

A surprising source of variance in both studies was the interaction between identity and cosmetics. This finding, indicating that cosmetics affected different faces differently, was analysed further to reveal that the more than attractive a face was initially, the less of an increase in attractiveness cosmetics conferred. While this is an intuitive finding, information technology has non been demonstrated before, and was especially pronounced in the set of supermodels where the result size of the interaction was almost as big as that of cosmetics itself. Cosmetics confer bonny patterns of colouration to faces, enhancing sex typical features in skin reflectance [18,twenty], every bit well as smoothing skin homogeneity and colour distribution [22,24,40]. Female person faces that are considered bonny tend to have lighter skin, darker eyes, and redder lips than the boilerplate female face [41], which are all correlates of attractiveness [20,21], and in a recent study, are colourations that are conferred to faces past cosmetics [18]. It may be that the more attractive faces (i.east., of supermodels) already possess the most attractive features that cosmetics tin alter, and so there is little modify in bewitchery after an application. That less attractive faces receive more than of an increase from cosmetics also has practical implications. By definition, the bulk of women will prevarication around boilerplate attractiveness, and so a significant number of women could receive a boost in attractiveness from cosmetics.

Nosotros also found that the perceived quantity of cosmetics applied to faces played almost no role in the perceived attractiveness of faces with cosmetics. Recent evidence has shown that faces with lighter makeup are perceived as more attractive than faces with heavier makeup [42], which is at odds with our findings here. However, that written report used different models for each cosmetics condition, conflating sources of cosmetics and identity variance, as well as using digitally applied cosmetics. While observers seem to find lighter cosmetics optimally bonny when given the option to vary the quantity [xvi], no study as of yet has systematically shown that lighter cosmetics are optimally bonny for a given face. Our measurements here, as well equally previous data [15], seem to advise quantity does non play a large role in perceptions of attractiveness with cosmetics.

Combining paradigm sets from previous enquiry [xv] with the findings from Study 1 revealed that, while the models from Report 1 were slightly more than attractive than the models from the previous study, they were rated every bit significantly more attractive with cosmetics. After considering the similarity of designs and full variability across all studies (both hither and in [15]), we compared the effect sizes of identity and cosmetics directly. Variability due to attractiveness between individuals (identity) was smaller among models and supermodels compared to university students, every bit predicted, but the issue size of cosmetics was noticeably larger for professionally-applied cosmetics. Notwithstanding, it is important to note that the sample sizes of models differed, and larger sample sizes might as well result in greater between-person variability.

These findings have relevance for investigating the furnishings of cosmetics on social perceptions. In that location at present exist estimates of the effect size of cosmetics when they are self-practical [xv], and when they are applied professionally. In previous work [fifteen], cosmetics explained but 2% of the variation in attractiveness, while the finding from a sample of models showed cosmetics explained 33% of the variation in attractiveness. This study demonstrated larger effect sizes of cosmetics when straight compared to previous research [fifteen], though the studies used different sets of faces, and it is important to annotation that any effect size estimate calculated is ultimately based on the context of the enquiry, and should be interpreted within this context [43]. Nonetheless, the variances in the current and previous enquiry are very similar, and the blueprint of the studies is identical, significant straight comparisons are valid and appropriate.

The literature examining the result of cosmetics on social perceptions has, for the almost function, used models with professionally-applied cosmetics in laboratory studies [9,ten,17,xxx,31] too as field experiments [12,13,44,45]. With our comparing of the result size of cosmetics under both self-applied and professionally—aaaapplied weather condition, it seems possible that some of the effects of cosmetics observed in the literature may be inflated. Farther, women report higher self-confidence and appoint in more social activities after a professional person makeover [46] and this increase in self-confidence may translate into slight expression or postural differences in images, which could represent an additional inside-person boost in bewitchery due to cosmetics.

There are some caveats to the report. Images were obtained from various Internet sources, and so were not as constrained in lighting or emotional expression equally previous inquiry [15]. Written report 1 suffered less from this potential issue every bit images were nerveless from the same photographic session. As the images of supermodels with cosmetics were obtained from different sources, while the images of those women without cosmetics were obtained from the same source, the magnitude of the interaction between identity and cosmetics should be interpreted with caution. Still, given its presence in Study 1 with more controlled stimuli, we think it condom to conclude that cosmetics touch more attractive individuals to a lesser extent than others. Furthermore, that such an upshot was obtained in Study 2 with more than variable photographs could exist considered strong evidence. Since the images were more variable and cosmetics were confounded with variations in lighting (both considered noise in the electric current study), it seems likely an effect would exist obtained under stricter conditions.

In that location now exists convincing prove that alterations to within-person facial appearance via cosmetics, whether cocky-practical or professionally-practical, do not overcome between-person variability in attractiveness due to unproblematic identity. Facial bewitchery is, to an extent, more about what you lot have, rather than what yous practice with it. However, we have uncovered here interesting caveats to this overarching and consistent finding. An increased skill level in applying cosmetics seems to offer a larger increase in attractiveness than self-practical cosmetics does—larger effects were clear when a professional person makeup artist applied cosmetics. Furthermore, nosotros take shown cosmetics affect faces of varying levels of bewitchery differently, particularly within a sample of faces with lower variation in attractiveness between individuals. More attractive individuals but have less to gain from using cosmetics. These findings have theoretical implications for attractiveness research. Cosmetics is perhaps the almost common form of modification of facial appearance, and we take shown that the currently reported literature, with its reliance on professionally-applied cosmetics, highlights an effect that does non seem achievable through everyday utilise.

How cosmetics affect attractiveness is a growing literature, and many studies apply professionally-applied cosmetics every bit a means to examine this change. Nosotros have shown that professionally-applied cosmetics seem to explain a larger proportion of variation in attractiveness judgements than self-practical cosmetics, a category which the vast majority of cosmetics users autumn nether. This could suggest an inflation of the effect of cosmetics in the current literature, with cosmetics increasing attractiveness beyond what is achievable through everyday means. Additionally, nosotros have illustrated that cosmetics impact women differently—more bonny women, particularly supermodels, proceeds less of a boost in attractiveness from cosmetics than practice less attractive women. Importantly, the outcome size of identity, or between-person variance in attractiveness, was larger than the consequence of cosmetics in both studies. We conclude that, when it comes to cosmetics, individual differences in facial appearance are ultimately more important than fifty-fifty a professional person application of cosmetics.

Supporting Information

S1 Dataset

Data from Study ane.

Each participant rated all 33 YouTube models, merely each model appeared in a randomly selected cosmetics condition. All conditions are stated in the data. Nosotros averaged across participants for each image, edifice a score for each identity nether both cosmetics conditions.

(XLS)

S2 Dataset

Data from Written report 2.

Each participant rated all 45 supermodels, just each model appeared in a randomly selected cosmetics status. All conditions are stated in the data. We averaged across participants for each image, building a score for each identity under both cosmetics conditions.

(XLS)

S3 Dataset

Information from the quantity raters in both studies.

Sheet 1 contains the quantity data from Study 1, and Sheet 2 contains the quantity data for Study 2. Participants compared each model without and with cosmetics, indicating how much cosmetics the faces were wearing.

(XLS)

Acknowledgments

The authors would like to thank Mila Mileva for assistance in collecting and preparing stimuli, testing participants, and providing advice on cosmetics looks. Thanks also go to Sarah Kramer for her help in participant testing and recruitment.

Funding Argument

The authors received no specific funding for this work.

Data Availability

All relevant data are inside the newspaper and its Supporting Information files.

References

1. Encephalon R. The decorated body. London: Hutchinson & Co; 1979. [Google Scholar]

2. Jablonski NG. Skin: A natural history. Berkeley: University of California Press; 2006. [Google Scholar]

3. Russell R. Why Cosmetics Work In: Adams RB, Ambady A, Nakayama K, Shimojo S, editors. The Science of Social Vision. Oxford: Oxford University Press; 2010. pp. 186–204. x.1093/acprof:oso/9780195333176.003.0011 [CrossRef] [Google Scholar]

4. Etcoff N. Survival of the prettiest: The science of beauty. New York: Ballast Books; 1999. [Google Scholar]

5. Rossi Eastward, Prlic A, Hoffman R. A study of the European cosmetics industry [Internet]. 2007. Available: ec.europa.european union/DocsRoom/documents/13125/attachments/ane/…/pdf.

6. Korichi R, Pelle-de-Queral D, Gazano G, Aubert A. Why women use makeup: implication of psychological traits in makeup functions. J Cosmet Sci. 2008;59: 127–137. [PubMed] [Google Scholar]

7. Miller LC, Cox CL. For appearances' sake: Public self-consciousness and makeup utilise. Pers Soc Psychol Bull. 1982;8: 748–751. [Google Scholar]

viii. Robertson J, Fieldman G, Hussey T. "Who wears cosmetics?" Private differences and their human relationship with cosmetic usage. Individ Differ Res. 2008;half dozen: 38–56. [Google Scholar]

9. Nash R, Fieldman G, Hussey T, Lévêque J-L, Pineau P. Cosmetics: They influence more than Caucasian female facial attractiveness. J Appl Soc Psychol. 2006;36: 493–504. 10.1111/j.0021-9029.2006.00016.x [CrossRef] [Google Scholar]

x. Etcoff N, Stock S, Haley LE, Vickery SA, Business firm DM. Cosmetics as a feature of the extended human phenotype: modulation of the perception of biologically important facial signals. Suzuki S, editor. PLoS I. 2011;6: e25656 x.1371/journal.pone.0025656 [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

11. Mileva VR, Jones AL, Russell R, Little Air conditioning. Sex differences in the perceived authorisation and prestige of women with and without cosmetics. Perception. 2016; 0301006616652053. x.1177/0301006616652053 [PubMed] [CrossRef] [Google Scholar]

12. Guéguen N, Jacob C. Enhanced female attractiveness with utilise of cosmetics and male person tipping beliefs in restaurants. J Cosmet Sci. 2011;62: 283–290. [PubMed] [Google Scholar]

13. Guéguen Northward. The effects of women's cosmetics on men's approach: An evaluation in a bar. Northward Am J Psychol. 2008;10: 221–228. [Google Scholar]

14. Cash TF, Dawson Yard, Davis P, Bowen Thou, Galumbeck C. Effects of cosmetics use on the physical attractiveness and body image of American college women. J Soc Psychol. 1989;129: 349–355. 10.1080/00224545.1989.9712051 [CrossRef] [Google Scholar]

15. Jones AL, Kramer RSS. Facial cosmetics accept footling outcome on attractiveness judgments compared with identity. Perception. 2015;44: 79–86. ten.1068/p7904 [PubMed] [CrossRef] [Google Scholar]

16. Jones AL, Kramer RSS, Ward R. Miscalibrations in judgements of bewitchery with cosmetics. Q J Exp Psychol. 2014;67: 2060–2068. 10.1080/17470218.2014.908932 [PubMed] [CrossRef] [Google Scholar]

17. Mulhern R, Fieldman G, Hussey T, Lévêque J-L, Pineau P. Do cosmetics heighten female person Caucasian facial attractiveness? Int J Cosmet Sci. 2003;25: 199–205. 10.1046/j.1467-2494.2003.00188.x [PubMed] [CrossRef] [Google Scholar]

eighteen. Jones AL, Russell R, Ward R. Cosmetics alter biologically-based factors of beauty: Evidence from facial contrast. Evol Psychol. 2015;thirteen: 210–229. 10.1177/147470491501300113 [PubMed] [CrossRef] [Google Scholar]

nineteen. Russell R. Sex, beauty, and the relative luminance of facial features. Perception. 2003;32: 1093–1107. 10.1068/p5101 [PubMed] [CrossRef] [Google Scholar]

twenty. Russell R. A sex activity difference in facial contrast and its exaggeration by cosmetics. Perception. 2009;38: 1211–1219. x.1068/p6331 [PubMed] [CrossRef] [Google Scholar]

21. Stephen ID, McKeegan AM. Lip color affects perceived sex typicality and bewitchery of human being faces. Perception. 2010;39: 1104–1110. 10.1068/p6730 [PubMed] [CrossRef] [Google Scholar]

22. Fink B, Grammer K, Matts P. Visible skin color distribution plays a office in the perception of age, attractiveness, and health in female person faces. Evol Hum Behav. 2006;27: 433–442. 10.1016/j.evolhumbehav.2006.08.007 [CrossRef] [Google Scholar]

23. Samson N, Fink B, Matts PJ. Visible peel condition and perception of human facial appearance. Int J Cosmet Sci. 2010;32: 167–184. 10.1111/j.1468-2494.2009.00535.x [PubMed] [CrossRef] [Google Scholar]

24. Jones AL, Porcheron A, Sweda JR, Morizot F, Russell R. Coloration in unlike areas of facial skin is a cue to health: The role of cheek redness and periorbital luminance in wellness perception. Body Image. 2016;17: 57–66. 10.1016/j.bodyim.2016.02.001 [PubMed] [CrossRef] [Google Scholar]

25. Porcheron A, Mauger Eastward, Russell R. Aspects of facial dissimilarity decrease with historic period and are cues for historic period perception. de Beeck HPO, editor. PLoS ONE. 2013;viii: e57985 ten.1371/periodical.pone.0057985 [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

26. Kościński Grand. Hand bewitchery—its determinants and associations with facial attractiveness. Behav Ecol. 2012;23: 334–342. 10.1093/beheco/arr190 [CrossRef] [Google Scholar]

27. Rhodes G. The evolutionary psychology of facial beauty. Annu Rev Psychol. 2006;57: 199–226. ten.1146/annurev.psych.57.102904.190208 [PubMed] [CrossRef] [Google Scholar]

28. Penton-Voak I. In retreat from nature? Successes and concerns in Darwinian approaches to facial attractiveness. J Evol Psychol. 2011;9: 173–193. 10.1556/JEP.nine.2011.2.five [CrossRef] [Google Scholar]

29. Morrison ER, Morris PH, Bard KA. The stability of facial bewitchery: Is it what you lot've got or what you lot do with it? J Nonverbal Behav. 2013;37: 59–67. 10.1007/s10919-013-0145-1 [CrossRef] [Google Scholar]

30. Cox CL, Glick WH. Resume evaluations and cosmetics utilise: When more than is not better. Sex Roles. 1986;14: 51–58. 10.1007/BF00287847 [CrossRef] [Google Scholar]

31. Huguet P, Croizet J-C, Richetin J. Is "what has been cared for" necessarily skilful? Further evidence for the negative affect of cosmetics use on impression germination. J Appl Soc Psychol. 2004;34: 1752–1771. x.1111/j.1559-1816.2004.tb02796.ten [CrossRef] [Google Scholar]

32. Tiddeman B, Burt DM, Perrett D. Prototyping and transforming facial textures for perception inquiry. IEEE Comput Graph Appl. 2001;21: 42–l. ten.1109/38.946630 [CrossRef] [Google Scholar]

33. Peirce JW. PsychoPy—psychophysics software in Python. J Neurosci Methods. 2007;162: 8–13. 10.1016/j.jneumeth.2006.xi.017 [PMC complimentary commodity] [PubMed] [CrossRef] [Google Scholar]

34. Coetzee V, Greeff JM, Stephen ID, Perrett DI. Cross-cultural agreement in facial attractiveness preferences: The role of ethnicity and gender. McCormick C, editor. PLoS Ane. 2014;ix: e99629 10.1371/journal.pone.0099629 [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

35. Howell DC. Statistical methods for psychology. 4th ed Belmont, CA: Wadsworth; 1997. [Google Scholar]

36. Cross JF, Cantankerous J. Age, sex activity, race, and the perception of facial dazzler. Dev Psychol. 1971;5: 433 10.1037/h0031591 [CrossRef] [Google Scholar]

37. Wickham LH, Morris PE. Bewitchery, distinctiveness, and recognition of faces: Attractive faces can be typical or distinctive but are not improve recognized. Am J Psychol. 2003;116: 455 10.2307/1423503 [PubMed] [CrossRef] [Google Scholar]

38. Lakens D. Calculating and reporting event sizes to facilitate cumulative science: a applied primer for t-tests and ANOVAs. Front Psychol. 2013;4: 863 10.3389/fpsyg.2013.00863 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

39. Jenkins R, White D, Van Montfort Ten, Burton AM. Variability in photos of the same face. Cognition. 2011;121: 313–323. 10.1016/j.cognition.2011.08.001 [PubMed] [CrossRef] [Google Scholar]

xl. Samson N, Fink B, Matts PJ. Visible skin condition and perception of homo facial appearance. Int J Cosmet Sci. 2010;32: 167–184. 10.1111/j.1468-2494.2009.00535.x [PubMed] [CrossRef] [Google Scholar]

41. Said CP, Todorov A. A statistical model of facial attractiveness. Psychol Sci. 2011;22: 1183–1190. 10.1177/0956797611419169 [PubMed] [CrossRef] [Google Scholar]

42. Tagai Yard, Ohtaka H, Nittono H. Faces with light makeup are improve recognized than faces with with heavy makeup. Forepart Psychol. 2016;7 x.3389/fpsyg.2016.00226 [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

43. Fritz CO, Morris PE, Richler JJ. Effect size estimates: current use, calculations, and interpretation. J Exp Psychol Gen. 2012;141: ii–eighteen. ten.1037/a0024338 [PubMed] [CrossRef] [Google Scholar]

44. Guéguen Due north, Jacob C. Lipstick and tipping beliefs: When red lipstick heighten waitresses tips. Int J Hosp Manag. 2012;31: 1333–1335. 10.1016/j.ijhm.2012.03.012 [CrossRef] [Google Scholar]

45. Jacob C, Guéguen Due north, Boulbry G, Ardiccioni R. Waitresses' facial cosmetics and tipping: A field experiment. Int J Hosp Manag. 2010;29: 188–190. 10.1016/j.ijhm.2009.04.003 [CrossRef] [Google Scholar]

46. Holme SA, Beattie PE, Fleming CJ. Corrective camouflage advice improves quality of life. Br J Dermatol. 2002;147: 946–949. x.1046/j.1365-2133.2002.04900.10 [PubMed] [CrossRef] [Google Scholar]

---

Articles from PLoS ONE are provided here courtesy of Public Library of Science

---

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5058481/

Posted by: cattplithenewark.blogspot.com

Share this post