Gender Differences in Preference for Product Design

By

Ellie Taylor

2008-2009

A PSYC3170 Major Project Supervised by

Dr Steve Westerman and Dr Ed Sutherland

A dissertation submitted in partial fulfilment of the requirements of the degree of

BSc (International) Psychology

And in agreement with the University of Leeds’
Declaration of Academic Integrity

[pic]

Institute of Psychological Sciences
University of Leeds

Acknowledgements

I would like to personally thank Steve Westerman, Ed Sutherland and Peter Gardner for the help and time they gave me to complete this project. Without their constant support it would have been much harder to produce.

A special thanks goes to Steve Westerman for creating the computer programme used, and guidance with statistics.

CONTENTS
Title Page…………………………………………………………………………………………………………………………………………………….1
Acknowledgements……………………………………………………………………………………………………………………………………...2
Contents Page……………………………………………………………………………………………………………………………………………..3
Abstract…………………………………………………………………………………………………………………….…..4
1. Introduction………………………………………………………………………………………………………………..5 1. How are Aesthetic Preferences Formed………………………………………………………………………….……………6 1. Previous Experience…………………………………………………………………………………………….……………..6 2. Physiological Feelings and Threat Perception……..……………………………………………………….………..6 3. Evolution…………………………………………………………………………………………………………………….…….7 4. Processing…………………………………………………………………………………………………………………….…..7 2. Aesthetic Preferences………………………………….…………………………………………………………………….………8 1. Contour…………………………………………………………………………………………………………….……………..9 2. Graphic………………………………………………………………………………………………………………….…………9 3. Complexity…………………………………………………………………………………………………………..…………10 4. Symmetry…………………………………………………………………………………………………………………….…10 5. Attention Grabbing………………………………………………………………………………………………….……….11 6. Typicality………………………………………………..……………………………………………………………….……..11 3. Gender Differences In Aesthetic Preference………..…………………………………………………………….………12 1. Shape Preference……………………………………………………………………………………………………..……..12 2. Difference in Designing…………………………………………………………………………………………………….12 3. Physiological Responses and Threat Perception…………………………………………………………..………13
1.4 Aims and Hypotheses……………………………………………………………………………………………………….………..13
2. Method…………………………………………………………………………………………………………………….15
2.1 Participants…………………………………………………………………………………………………………………………….……..15
2.2 Materials………………………………………………………………………………………………………………………………..……..15 2.2.1 Participant Documents………………………………………………….......................................………..……….15 2.2.2 Measures…………………………………………………………………………................................................…..15 2.2.3 Computer Programme…………………………………………………................................................…………16
2.3 Stimuli…………………………………………………………………………………………………………………………………..………16
2.4 Design………………………………………………………………………………………………………………………………………..…17
2.5 Ethical Considerations……………………………………………………………………………………………………………….……17
2.6 Procedure……………………………………………………………………………………………………………………………………..18
2.7 Data Analysis…………………………………………………………………………………………………………………………………19
3. Results ……………………………………………………………………………………………………………..…….20
3.1 General Response Ratings……………………………………………………………………………………………………………….21 3.1.1 Aesthetic Preference……………………………………………………………………………………………………………..21 3.1.2 Attention Grabbing…………………………………………………………………………………………………...…………..22 3.1.3 Purchase Likelihood………………………………………………………………………………………………………….……22 3.1.4 Typicality…………………………………………………………………………………………………………………....……….24 3.1.5 Usability………………………………………………………………………………………………………………………..……..24
3.2 Affective Responses…………………………………………………………………………………………………………….………….26 3.2.1 Anxious…………………………………………………………………………………………………………………….………….26 3.2.2 Tense……………………………………………………………………………………………………………………….………….27 3.2.3 Relaxed……………………………………………………………………………………………………………………..…………28
4. Discussion………………………………………………………………………………………………………………..30
4.1 General Responses…………………………………………………………………………………………………………………………30
4.2 Affective Responses………………………………………………………………………………………………………………..………33
4.3 Final Conclusions and Future Directions……………………………………………………………………………………………35

References…………………………………………………………………………………………………………………………………………………37
Appendices………………………………………………………………………………………………………………………………………………..40
A. Ethics Documents…………………………………………………………………………………………………………………..……..40
B. Supervision Diary…………………………………………………………………………………………………………..………………48
C. Screen Shots of Computer Programme…………………………………………………………………………………………….49
D. SPSS Data Output………………………………………………………………………………………………………….………………52
E. Raw Data……………………………………………………………………………………………………………..………………………75

ABSTRACT

It has been found that in general rounded shapes are preferred over more angular shapes. Bar and Neta (2006; 2007) found participants picked smooth products over sharp products when asked to make binary like/dislike judgements. They further suggested that this finding was due to sharp angles evoking a feeling of threat amongst people, which triggers a negative bias. They supported this by finding sharp contours elevated arousal within the amygdala due to an implicit association of sharp features with danger. Furthermore people judge rounder faces as more attractive than angular ones (Zebrowitz 1997). Research has also begun to touch upon gender difference in product design preference. Xue and Yen (2007) identified that males tended to select fragrance bottles of regular geometric forms, whereas females chose fluid more curvy bottles. Such gender preference have even been found in young children, demonstrated by girls tending to create drawings of rounded objects and boys of sharp lines and angles (Franck and Rosen 1949).
The aim of this study was to further examine if there is a significant gender difference for product design, observing whether each gender had a particular preference for bottle shape and graphics used. There were 50 male and 56 female participants with an average age of 21 years and 5 months. A computer task was used which presented participants with 4 different bottles, differing in shape, contour and graphics, they then had to rate the bottles on 57 constructs (e.g. how beautiful, attention grabbing, functional etc) using a Likert Scale. It was found that overall participants did have an aesthetic preference for rounded contours and graphics compared to sharp, however there were no gender difference found. It was also found that sharp contours and graphics did cause participants to make higher conscious ratings of feeling more anxious and tense when observing these products. The results of this research will be useful for product design companies, as they can use them to help design their products and effectively target their specific markets.

1. INTRODUCTION

The general public spends £144 million per week within the retail industry, with the average family spending £454 per week. However with the poor economic climate, statistics show that the average family now has £13 less a week available to spend in 2009 compared to 2008(National Statistics Online). With so many products available in such a competitive market, what makes a consumer purchase one product over another? Possibly one of the key factors which persuade consumers to purchase certain products are the aesthetics and the way it is presented. People make judgments on products based on factors such as symmetry, prototypicality, contrast, complexity, and perceptual fluency (Reber, Schwarz, & Winkielman, 2004). With today’s economic climate in a poor state, people have reined in on their spending and companies will have to try even harder to sell their products. Therefore it is important that each product is eye catching and attractive to its target audience. Understanding the differences in preference between your target audiences is vital in order to market your product towards them. What is aesthetically pleasing has been examined for centuries now, with a number of views emerging from the investigation into its roots. The objective view suggests that beauty is the properties of an object that produce a pleasurable feeling within the observer. This view suggests we can examine what properties affect aesthetic preference and alter them to make something more beautiful. However the subjective view believes there are no set properties that make an object beautiful or not, it is simply ‘beauty is within the eye of the beholder’ and will vary between each person. Finally there is an interactionist view which believes beauty emerges from the way the observer and object relate, e.g. the processing experience of an object can affect the observers’ feelings towards the object (Reber 2004). This present study will discuss the objective view and examine what properties are perceived as more aesthetically pleasing; particularly examining contours e.g. sharp or round. We also wish to highlight the interactionist view and see if the way a product makes someone feel e.g. the way they relate to it, affects their preference for it, particularly examining threat perception in relation to contours.
1.1 How are Aesthetic Preferences Formed?
Bargh (1982) suggests that although one is aware of the properties of an object, they are not actually aware how these properties influence their preferences for the object. Thus suggesting that aesthetic preference is formed in a non-conscious manner. As first impressions are formed so quickly (Ambady Bernieri & Richeson 2000) this further supports the notions that preferences are formed non-consciously, as the process occurs too quickly for it to be consciously performed. It has been suggested that previous experience, threat perception, evolution and processing fluency are some of the possible factors which influence how our preferences are formed.
1.1.1 Previous Experience
Liu (2003) highlights that it is not simply the features of the objects which influence judgments, but also the objects symbolic meaning which sways preference. Each individual will have differing memories and mental associations related to the object which are evoked upon observation. If such associations are positive e.g. a particular shampoo bottle reminds them of their Mum washing their hair when they were younger, then they are more likely to have a positive bias towards the object and potentially find it more aesthetically pleasing. However if the object evokes negative memories e.g. a particular bleach bottle reminds them of when they were young and drank from it and were very ill, then a negative bias may be formed, and thus it is less likely to be aesthetically pleasing.

1.1.2 Physiological Feelings and Threat Perception
Bar and Neta (2007) discovered that people had a liking for rounded contours over sharp contours, however they went a step further and examined why people held such a preference by examining brain activity when viewing sharp and round contours. It was found that the left and right amygdala had significantly greater activation when the participant was viewing sharp contours. The amygdala is related to emotional responses, particularly fear, thus it is suggested that sharp contours elicit a fear response which in turn causes a negative bias, due to the perception of threat. Research discussed in relation to complexity and symmetry also showed a similar pattern, that when a strong physiological reaction was elicited in response to viewing complex or asymmetrical patterns, a negative bias was created towards it.
Aronoff et al (1992) had previously proposed that when human facial expressions or movements contain sharp features (baring teeth, bent elbows) it conveys a sense of threat; where as round features convey warmth. This idea may be because these kinds of features usually occur in fighting e.g. an elbow bent ready to punch, or a leg bent at the knee preparing to kick out. These feelings, of sharp contours on the human body evoking a sense of threat, has been transferred onto objects too, thus angular features are not liked as much as rounded features. Cannon (1915) found that when a person feels threatened, it triggers a fight or flight response within the body, which causes an increase in arousal levels. Berlynes’ Arousal Model suggested that there is an inverted U shape between arousal and liking. Therefore very little or very extreme arousal levels cause liking to be very low, however when arousal is at a medium level, liking is at its highest. This model ties in well with threat perception, as threat increase arousal to a very high level, thus what ever caused such threat, thus arousal, is going to be disliked, in this case the sharp or angular contours.

1.1.3 Evolution
As Berlynes’ Arousal Model suggests liking, and potentially aesthetic preference, is routed within our physiology and thus possibly our evolutionary past. We may have developed such physiological responses to objects through adaptive change to help aid survival. This is highlighted by the fact that people have a preference for attractive faces, this is due to such faces indicating good health and good genes which aids procreation. Zebrowitz (1997) found people judged rounded faces to be more attractive than angular faces. Thus rounded contours are preferred; this particular preference in faces may also transfer for objects with rounded contours.

1.1.4 Processing
If we draw upon the interactionist view, the perception of beauty is created through the processing of the object by the observer. Reber et al (2004) suggested that the more fluently the observer can process an object, the more positive their aesthetic response is e.g. if easily processed they will have an aesthetic preference for the object. In order to process an item fluently we evaluate constructs such as contour, symmetry and complexity for example. An example of this idea is used in the preference for prototypes model, which suggests that we have an aesthetic preference for objects that are typical for their class. Whitfield and Slatter (1979) tested this model using different kinds of chairs, some designs more prototypical than others. It was found that the chair designs most typical to what people expect a chair to look like were rated as most liked, and thus aesthetically preferred. This model suggests that ‘typical’ objects are preferred because they are more easily classed, e.g. processed more fluently. This idea of fluent processing increasing liking is also demonstrated by preference for attractive faces. Griffin (2007) found attractive faces to be processed more fluently than unattractive faces. She explained these findings in terms of the Averageness theory, that suggests attractive faces are said to be preferred because they are more average and closely resemble the rest of the population and thus are more familiar and hence more ‘protypical’. As suggested by the Prototype Theory, if an item is more typical of its class it is processed more fluently, and as Reber said fluent processing equals increased preference.

1.2 Aesthetic Preferences

We can draw upon previous experience, threat perception, evolution and processing to explain how our preferences are formed, but there has also been vast research into what our aesthetic preferences are. What is aesthetically pleasing has been examined for centuries now, with a number of views emerging from the investigation into its roots. The objective view suggests that beauty is the properties of an object that produce a pleasurable feeling within the observer. This view suggest we can examine what properties affect aesthetic preference and alter them to make something more beautiful. There are a number of factors which have an impact upon our judgment of product designs; these include type of contour, graphic, complexity, symmetry, attention grabbing properties and typicality.

1.2.1 Contour
Bar and Neta (2006) looked at preferences for contour. They used letters, patterns and objects of neutral valence but varied on contour, either sharp or rounded. It was found for all type of stimuli that rounded contours were preferred over sharp. Thus it was suggested rounded contours are an aesthetic preference for humans. Furthermore they found that preference was higher for real life object over patterns or letters. This finding suggests people have a preference for items when they are familiar. More recently this idea of preference for rounded contours was further supported by a study which looked at either angular or rounded polygon shapes(Silva 2009). It meant this study would not be affected by confounds such as familiarity or typicality. Thus further evidence was found for the basic property of contour affecting preference e.g. rounded is favored. Such preference for rounded contours has not only been found for objects and patterns, this idea has also been shown for human figure (Guthrie and Wiener 1966). It was suggested that when a human figure is made to look jagged i.e. shoulder, elbows and knees bent in an angular fashion, people rate this as aggressive which is a negative trait, thus they have a negative preference for such angular features. Furthermore Zebrowitz (1997) found that the aesthetic preference for rounded contours was also present when judging faces. People who had rounder faces were rated as more attractive and were liked more than those with angular faces. Overall evidence points towards a trend of rounded contours being preferential over angular contours for a variety of objects.

1.2.2 Graphic
The type of graphics featured on a product can also affect liking. Leder and Carbon (2005) presented participants with car interiors featuring either curved or straight shapes. It was shown that the most curved versions of interior design gained the highest attractiveness ratings. They suggested this finding was due to curvature causing increased positive responses due to curvature being associated with cuteness and beauty. This demonstrates that not only are the curved contours of a shape itself preferred but also the curved graphics within it help enhance liking.

1.2.3 Complexity
The complexity of a design can also enhance or reduce liking, Osborne and Farley (1970) presented famous paintings varying in complexity to participants. They found that more complex paintings were preferred. This suggests that aesthetic preference is found for complex designs. However Hekkert and Wieringen (1990) did not support such findings, upon presenting cubist paintings they found that beauty ratings had an inverted U relationship with beauty. Thus people did not like very simple or very complex paintings. Berlyne (1974) had previously developed an explanation as to why people preferred medium levels of complexity in designs. He suggested that complexity refers to the arousing potential of a stimulus. Upon presenting artificial patterns he found that people rated medium complex designs the highest as they did not arouse too much (which complex objects do) but do cause a suitable amount of arousal, unlike very plain boring designs. This suggests that people will prefer a design which causes some response within the body, but a strong physiological response causes a negative bias.

1.2.4 Symmetry
There is vast previous research which has found that symmetry is a very important factor when rating aesthetic preference for faces, with symmetrical faces being rated as more attractive than asymmetrical (Jone and Little 2003). Cardenas and Harris (2006) further supported the idea that symmetry makes for higher aesthetic preference. They took symmetrical and asymmetrical faces and had participants rate them for attractiveness, as expected the symmetrical faces were given higher attractiveness ratings. However when the symmetrical faces were painted with symmetrical or asymmetrical designs it altered aesthetic preference. The asymmetrical designs painted upon the symmetrical face lowered its attractiveness’ ratings. This suggests that asymmetrical faces and patterns are preferred less. Hambridge’s study of Greek art caused him to come to the conclusion that the beauty of Greek art was due to the use of symmetry. Krupinski and Locher (1988) found that when participants viewed asymmetrical art work, compared to symmetrical work, they had higher levels of arousal, as shown by skin conductance tests. This was correlated with higher reporting of preference for the symmetrical art work. This suggests that symmetry is preferred over asymmetry due to asymmetrical designs eliciting an uncomfortable physiological response. Interestingly though Washburn and Humphrey (2001) found that preference for symmetry can be affected by Art Training and education. When art students were compared against non art students it was found the art students had higher preference for non salient forms of symmetry and used less symmetrical designs when constructing patterns compared to non art students. This demonstrates that art training can alter preferences for symmetry.

1.2.5 Attention Grabbing
How attention grabbing a design is can impact upon how likely someone is to purchase it. For example if a design grabs someone’s attention they are more likely to pick it up to have a look and thus increase purchase likelihood. Bredart et al(2006) examined the attention grabbing properties of faces and suggested that the familiarity of ones own face was what made it attention grabbing. This shows that when we are regularly exposed to a design and it becomes familiar, it will grab our attention at a later date.

1.2.6 Typicality
Typicality of a product is how representative it is of a particular product. Smarz (2005) found that as typicality increased for a product so did ratings of liking and interest. Whitfield and Slatter (1979) further highlight the importance of typicality in aesthetic preference. They varied chair designs in terms of typicality and found that the more typical the design was the more it was liked. This suggests that when presented with designs that we view as prototypical for its category, we will have an aesthetic preference for it.

1.3 Gender Difference in Aesthetic Preference

Research shows clear patterns for peoples general preferences for design, but do such preferences truly apply to both males and females, or do they differ somewhat? Research has shown differences in the way males and females designs objects and their preferences for object shape, such differences may be explained by the difference in threat perception between the genders.

1.3.1 Shape Preference
Moss (1995) reviewed a number of papers looking into gender difference and found that there are gender differences for design preference. Xue and Yen (2007) further examined this by presenting both genders with differing fragrance bottles. Females chose less geometrical shapes, and were concerned with slimness and smoothness, using words such as curvy and neat in their descriptions. Where as males chose regular geometric shapes and focused on compact clean lines and the bottles overall structure. This demonstrates that women appeared to have a preference for more rounded designs, where as males didn’t focus so strongly on this and simply wanted a strong compact design.

1.3.2 Difference in Designing
Research has demonstrated that the genders have different preferences when designing products and such differences are evident even from a young age. Erikson (1973) looked at pre adolescents who were asked to build a tower from blocks. Even at a young age the girls built low circular structures and the boys created square towers pointing upwards. Majewski (1978) then also found that when children were presented with different children’s drawings, the boys preferred drawings which featured rectilinear forms and girls had a preference for circularity. Such differences and preferences for designs then continues with age shown by Franck and Rosen (1949), who found when asked to complete various shapes, men emphasized sharp lines and angles whereas women drew round or blunt lines. The male preference for straighter angular designs and women’s for curvature extends to their business cards as well. Moss (2003) presented participants with male and female business cards and asked them to describe the cards. Male cards were singled out as using aggressive straight lines and 3 dimensional qualities. Whereas female cards used round fluent shapes and soft surfaces. This research points towards female’s greater preference for rounded designs, whereas males prefer more angular designs.

1.3.3 Physiological Responses and Threat Perception
These differences in aesthetic preference between males and females may be due to their differing physiological responses to threatening stimuli, e.g. angular contours. When presented with a mild electrodermal stimulation (a threatening action) women, compared to men, showed an increase in subgenual anterior cingulated cortex activity when anticipating the shock (Butler et al 2005). This area is related to conflict detection, suggesting that women are more sensitive to picking up on conflict, thus threat. This demonstrates that women may have a stronger aversion to sharp contours because they elicit a stronger fear response compared to men. McClure et al (2004) presented women and men with a threatening stimuli and found women showed greater orbitofrontal cortex and amygdale activation than men. Both of these areas are implicated in emotion. This lends further support to the idea that women are more sensitive to threatening stimuli and thus a more negative bias will be formed. As Bar and Neta suggested sharp transitions in contour may convey a sense of threat, if women are more sensitive to such a threat then they will like sharp contours less than males do.

1.4 Aims and Hypotheses
Previous research has pointed to the idea that generally people have an aesthetic preference for rounded contours over sharp. This may be due to the fact that sharp transitions in contour convey a sense of threat and thus a negative bias is formed. However it has not been examined if this preference for rounded contours is as strong in males and females. As research into gender differences when processing threatening stimuli suggests women have an increased response to threat stimuli than males in terms of brain activity. This points towards the idea that females will have a greater preference for rounded contours than males. Therefore this study aims to examine the aesthetic preference for products designed with sharp or rounded contours and whether the preference differs between males and females. Therefore this study will test the hypothesis that ‘Females will have a greater preference for products with rounded contours rather than sharp, compared to males.’ The study will also examine the hypothesis that ‘Products with sharp contours will elicit a greater anxious response and lower relaxed response, than rounded contours, for females compared to males.’ Finally the study will test the hypothesis that ‘Females will have a greater preference for rounded products and thus will increase purchase likelihood’.

2. METHOD

2.1 Participants
One hundred and six participants (50 males and 56 females) were recruited by opportunity sampling via the university participant pool scheme. All were undergraduate students from University of Leeds and Leeds Metropolitan University. The samples mean age was 21 years and 5 months with a range of 18.6 to 29.5 years.

2.2 Materials
2.2.1 Participant Documents
All participants were given a standardized information sheet (appendix 1) which explained that they would be expected to rate 4 different products on a number of different constructs; it also gave an estimated time and ethical guidelines. Participants were also given a consent form (appendix 2) to sign to ensure they agreed to partake and understood they could withdraw at anytime. A written de-brief (appendix 3) was also provided after completion of the study, this gave the participant more detailed information on what we were looking at and the expected direction of the results.

2.2.2 Measures
The first part of the study was a personality questionnaire consisting of 50 items testing extraversion, agreeableness, conscientiousness, neuroticism and openness to experience. Participants answered this using a 5 point Likert scale where 1= very inaccurate and 5= very accurate.

The second part of the study was measuring participants’ ratings on the products typicality, ease of use, aesthetics, abstract associations, objective assessments and affective responses. These were also rated on a similar 5 point Likert scale where 1= definitely not and 5= definitely.

2.2.3 Computer Programme

The presentation program was written specifically for the purpose of the study in Microsoft Visual Basic, by Steve Westerman. The bottle designs were developed using a program called 3D Studio. (Appendix 6)

2.3 Stimuli

A total of 8 bottles were created and presented to participants. A total of 4 of these bottles were ‘bleach’ bottles and had the word ‘bleach’ written across the front. The remaining 4 bottles were ‘water’ bottles with ‘water’ written across the front.

The 4 bleach bottles and 4 water bottles had the same designs, but had differing words written across the front (water or bleach). These 4 designs were comprised of two bottles that had rounded contours and two that had sharp contours. Of the rounded contour bottles one featured graphics of a rounded nature and the other featured sharp graphics. Of the sharp contoured bottles, one featured rounded graphics and the other featured sharp graphics. These bottles were created using a graphic design programme.

[pic][pic][pic][pic] Bleach Bottles Designs

[pic][pic][pic][pic] Water Bottle Designs

2.4 Design
This study was designed to examine 4 independent variables, gender (male or female) contour (sharp or round) graphic (sharp or round) and product (bleach or water). As each of these variables had two levels it created a 2x2x2x2 mixed design. The design was mixed because contour and graphic were within subjects variables and sex and product were between subject variables. Therefore half the males and half the females viewed bleach bottles and the other half viewed water bottles. However all participants viewed both sharp and round contours and graphics.

The dependant variables that were being measured were purchase likelihood, typicality, functionality, attention grabbing, aesthetic preference, anxious, tense and relaxed ratings.
The experimental control implemented was counterbalancing of the product order, thus the 4 bottles were presented to each participant in a random fashion. Furthermore group assignment was counterbalanced too, so half males and half females were randomly assigned to either the bleach or water condition.

2.5 Ethical Considerations

This study was not examining any particularly sensitive issues; however the participants’ welfare still needed to be taken into consideration so the British Psychological Societies guidelines were followed as well as ethical approval being sought from the University of Leeds Ethics Committee (appendix 4). The use of deception was not necessary to gain results, so participants were fully informed of the details of the study prior to taking part via an information sheet. They were then informed that they had the right to withdraw from the study at any point without giving a reason and that their data was completely anonymous. They were then given the opportunity to ask any questions before finally signing a consent form.

2.6 Procedure
Participants first read the information sheet and ethics form and were given the chance to ask any questions before they signed a consent form agreeing to partake in the study. Participants were sat down in front of a computer, in the Psychology of Design Laboratory. The experimenter assigned them to either the bleach or water condition, then gave them a participant number and filled in their initials, age in years and months and gender. The experimenter then vacated the room to reduce any social desirability.

The participants began the study by reading a set of onscreen instruction relating to the personality section of the study. Participants then answered a personality questionnaire comprised of 50 elements, these were answered by inputting a rating of 1-5 that most accurately described them. This stage took about 15 minutes.

The next stage was the product design ratings, which began with another set of on screen instructions explaining what was expected. They were also told they could have a breather before beginning the next section.

The participants were then presented with 4 bottles of differing designs, one at a time. They had to answer 57 rating questions for each bottle, therefore there was 228 ratings to be made in total. This took about 30 minutes but participants were given a breather in between the presentation of each new bottle design.

Upon completion of the experiment, the participant was verbally thanked for partaking and given a debrief sheet. Finally they were given the chance to ask any other questions and reminded that they could withdraw their data at any time by contacting the experimenter.

2.7 Data Analysis
Data was entered into SPSS 14.0. ANOVA was carried out on the dependant variable attention grabbing and purchase likelihood. However typicality, usability and aesthetic preference were not analysed as an individual construct. For these a factor analysis was conducted first to examine clusters of correlating descriptors for each of these constructs. An ANOVA was then carried out on the clusters. (appendix 7)

3. RESULTS

This study aimed to identify whether males and females have differing preferences for products design. This general idea was measured by looking at the following constructs; aesthetic preference, attention grabbing, purchase likelihood, typicality and usability. We further looked at the conscious feelings (anxiousness, tension and relaxation) participants felt towards the products as this may affect liking. When analysing the attention grabbing, purchase likelihood, anxious, tense and relaxed constructs we used the data directly from the data file. However for typicality, usability and aesthetic preference a scale of constructs was used to measure the overall construct. The data collected for this report was part of a larger research project therefore not all the data is reported and analysed. All output can be seen in Appendix 7.
Table 1. The average means and standard deviation of all the constructs analysed.
| |Aesthetic |Attention |Purchase |Typicality |Usability |
| |Preference |Grabbing |Likelihood |Mean (S.d) |Mean |
| |Mean (S.d) |Mean (S.d) |Mean | |(S.d) |
| | | |(S.d) | | |
|Male |Graphic |Round |2.14 (0.12) |2.38 (0.14) |2.68 (0.12) |2.5 (0.09) |3.51 (0.09) |
| | |Sharp |2.02 (0.12) |2.46 (0.13) |2.6 (0.12) |2.6 (0.09) |3.38 (0.09) |
| |Contour |Round |2.27 (0.12) |2.24 (0.13) |2.66 (0.14) |2.3 (0.10) |3.64 (0.10) |
| | |Sharp |1.89 (0.09) |2.6 (0.16) |2.62 (0.12) |2.8 (0.10) |3.25 (0.09) |
|Female |Graphic |Round |2.41 (0.12) |2.61 (0.13) |3.18 (0.12) |2.45 (0.09) |3.57 (0.08) |
| | |Sharp |2.12 (0.11) |2.6 (0.13) |2.85 (0.12) |2.51 (0.09) |3.49 (0.08) |
| |Contour |Round |2.21 (0.11) |2.56 (0.12) |3.35 (0.14) |2.13 (0.9) |3.87 (0.1) |
| | |Sharp |1.92 (0.09) |2.65 (0.15) |2.68 (0.11) |2.83 (0.1) |3.18 (0.09) |

Five 2x2x2x2 ANOVAS were conducted, there were two levels for each variable, graphic(sharp-round) contour (sharp round) gender (male-female) and product (bleach-water).
3.1 General Response Ratings
3.1.1 Aesthetic Preference
The Aesthetic Preference construct was created using a scale which combined beautiful, aesthetically pleasing and like the look ratings. An ANOVA was then carried out on the overall construct. A significant main effect was found for graphic F(1,105)= 10.68, p