Author Names: Jennifer Dow (BSc (Hons) Animal Behaviour and Welfare) and Sienna Taylor
The domestic cat (Felis catus) is a popular companion animal due to their friendly, yet independent nature. An estimated 65,000 cats end up in UK shelters a year, with 60% of cat shelters reporting that they reach full capacity all year round. If adoption rates are low within full capacity shelters, it can often result in cats spending long periods of time in shelter confinement which can cause prolonged fear and anxiety. This can cause chronic stress which in turn impacts upon cats’ normal behavioural and biological functioning, and therefore their welfare. Thus, it is vital to reduce the amount of time that cats spend in shelters in order to reduce the occurrence of chronic stress. In order to reduce the amount of time it takes for cats to be adopted, factors impacting upon cat adoption must be investigated. Numerous studies have investigated the impacts of environmental factors, however, little research has looked directly into the impact that cat coat colour has on cat adoption. The aim of this study was to assess whether cat coat colour affects the length of time taken to be adopted from a rescue shelter using archival research. Adoption records from 867 cats homed in RSPCA Birmingham Animal Shelter from the years 2012- 2014 were used. In total, five groups were looked at: Black, Black and White, Brown/ Brown Tabby, Ginger/ Tortoiseshell and White/ Mainly White cats. A combined Kruskal- Wallis and Dunns- Bonferroni test was used to test for any significant differences between coat colour and length of stay. It was found that there was a significant difference between coat colour and length of time taken to be adopted (p= <0.001). Black and white cats took the longest to be adopted in comparison to the other four groups. Whilst the difference was not significant, white/ mainly white cats were found to take the least amount of time to be adopted. In addition to this, a questionnaire was created using SurveyMonkey in order to identify whether the UK public have any preference on cat coat colour. Ginger/ tortoiseshell cats were found to be the most preferred, whereas white/ mainly white cats were found to be the least preferred by participants. Thus conflicting results were found between the adoption record analysis and the questionnaire. Preferences from the questionnaire were found to be based on preconceived information that colour directly related to behaviour, participants believing that certain colours are associated with health issues, participants showing less preference for cats prone to looking ‘dirty’ and preferences related to early childhood experiences. Further research should incorporate a larger sample size in order to further investigate trends and increase the generalisability of results, as well as reviewing the coat colour categories to avoid participant bias.
The domestic cat (Felis catus) has been living alongside humans for thousands of years and make for popular companion animals due to their friendly, yet independent nature (Turner and Bateson, 2014). In 2015, the pet cat population in the UK was 7.4 million (RSPCA, 2016). From this population, an estimated 65,000 cats end up in UK shelters a year, with 60% of cat shelters reporting that they are at full capacity all year round (Clark et al, 2012). If adoption rates are low within full capacity shelters, it can often result in cats spending long periods of time in shelter confinement or even euthanasia in some cases (Gourkow, 2001). Spending long periods of time in shelter conditions can be extremely stressful for cats due to novel smells and sounds, having close proximity with unfamiliar conspecifics and routine handling by staff (Kry and Casey, 2007; Rochlitz, 2007). Studies have found that the longer a cat spends in a shelter, the more chance it will experience chronic stress, which may negatively impact the cat’s physical, mental and social health, therefore impacting the cat’s welfare (Rodan and Heath, 2016; Kogan et al, 2013). Animal welfare can be defined as ‘the state of an animal as it attempts to cope with its environment’ (Broom, 1991). However, this definition does not define what constitutes good or bad welfare and so a more detailed explanation was provided by Webster (2005). They stated: ‘Is the animal living a natural life?, Is the animal fit and healthy? And Is the animal happy?’ (Webster, 2005). Webster (2005) suggested that this definition may be better suited, as each premise can be seen as necessary to establish a sufficient view of animal welfare (Webster, 2005). The Five Freedoms also define categories in which welfare could be promoted and ideally achieved (Webster, 2005). The freedoms are as follows: Freedom from thirst, hunger and malnutrition, freedom from discomfort, freedom from pain, injury and disease, freedom from fear and distress and freedom to express normal behaviour (Webster, 2005). If an animal does not have the ability to express these freedoms within reason, it is said to have poor welfare (Webster, 2005). Shelter life may typically impact on an animal’s freedom from fear and distress and freedom to express normal behaviour, as shelter environments have been reported to elicit fear and anxiety in cats (Holmes, 1997; Holmes, 1993). Prolonged fear and anxiety can cause chronic stress, which in turn impact on cats’ normal behavioural and biological functioning (Rodan and Heath, 2016; Holmes, 1997; Holmes, 1993).
1.1 Behavioural and Biological Impacts of Stress
The behavioural signs of stress that cats typically show are the inhibition of normal maintenance behaviour such as grooming, defensive aggression and the inhibition of play, consequently impacting on the cat’s freedom to express normal behaviour (Bradshaw et al, 2012). Additionally, the expression of these behaviours have been associated with depression in existing literature (Pederson and Pratt, 1991). As well as this, tense muscle tone, crouching posture and pupil dilation may all also be indicative of stress in cats (McCobb et al, 2005). In addition to stress-related behaviours, shelter environments have also been reported to evoke both fear and anxiety in cats (Holmes, 1997; Holmes, 1993). Holmes (1993) defined anxiety as an emotional state caused by the absence of stimuli such as a human companion or familiar objects to which the cat is attached. Holmes (1997) also defined fear as a motivational state provoked by specific external stimuli which promote avoidance, defensiveness and escape behaviour. Prolonged fear occurring at a constant, elevated level is believed to lead to chronic anxiety, depression and neurosis which significantly impact upon the cat’s normal functioning and therefore welfare (Morton, 1998). Visual signs of effects on normal functioning include heightened or reduced elimination behaviour, vocalisations, appetite, grooming habits and physical activity, all of which can be seen in cats displaying anxiety- disorders (Little, 2012).
Shelter life can also affect cats’ biological functioning due to chronic stress which greatly impacts welfare (Rodan and Heath, 2016). For example, Carlstead et al (1993) investigated the effects of an unpredictable handling and husbandry routine on eight ‘stressed cats’ within a confined, laboratory environment. Their routine consisted of irregular feeding, cleaning and the absence of human attention. Carlstead et al (1993) found that urinary cortisol concentrations were consistently elevated in the chronically stressed cats due to the change of routine, in addition to active exploratory and play behaviours being suppressed. Carlstead et al (1993) concluded that a constant elevated level of urinary cortisol could eventually have wider implications such as compromised reproduction, which in turn alters the cats’ biological function. Furthermore, it was found that when a stressor was related to cat isolation or confinement, the gastrointestinal system was impacted, affecting the cats biological functioning (Rodan and Heath, 2016). These effects included intermittent diarrhoea, vomiting and decreased appetite (Rodan and Heath, 2016). For cats living in shelters, changes in handling and husbandry routines as well as spatial and social restriction may be unavoidable (Rodan and Heath, 2016). Therefore, cats that spend longer durations within shelters due to low adoption rates, are more likely to become chronically stressed due to longer exposure to such chronic stressors (Rodan and Heath, 2016; Kogan et al, 2013).
It is therefore vital to increase shelter cats’ adoptability so that the amount of time that a cat spends at a shelter is reduced, in order to diminish the occurrence of stress and fear- induced behaviours (Bradshaw et al, 2012; Morton, 1998; Carlstead et al, 1993). Stress and fear- induced behaviours may negatively affect cats’ ability to express normal behaviour and biological functioning, which in turn affects their welfare (Bradshaw et al, 2012; Morton, 1998; Carlstead et al, 1993). To be able to achieve this, factors that affect shelter cats’ adoptability must be explored (Bradshaw et al, 2012).
1.2 Factors Affecting Cat Adoptability
There are a number of studies that have explored the effect that housing, cage enrichment, age, personality and playfulness have on cats’ adoptability from shelters. For example, Weiss et al (2012) found that cats’ health and behaviour were important factors prior to adoption, whereas Fantuzzi et al (2010) found that cats with a greater number of toys and who were housed at eye level were adopted more readily (Fantuzzi et al, 2010). In addition to this, Gourkow and Fraser (2006) concluded that a combination of positive handling and environments enriched by either communal housing or improved cage design could reduce shelter cats’ fearfulness, thus making the cats better candidates for early adoption, as relaxed and friendly behaviour may be more attractive to adopters (Gourkow and Fraser, 2006), These findings can be used to change factors such as cage height, increasing interaction time when adopters are present and providing environmental enrichment in order to increase shelter cats adoptability (Fantuzzi et al, 2010; Gourkow and Fraser, 2006).
Whilst environmental factors that impact cat adoption have been investigated in detail, there is limited research assessing the effect that a cat’s coat colour has on its adoptability. A study conducted in America by Delgado, Munera and Reevy (2012) found that people may perceive cat coat colour as a factor that contributes to the personality of cats. For example, their survey found that orange cats were considered to be relatively high in friendliness, whereas white cats were rated as shy, more aloof and calm (Delgado, Munera and Reevy, 2012). These differences may be due to the fact that Garfield (a ginger cat) features in a popular American TV show, thus displaying an orange cat as a friendly companion, whereas a popular cat food advertisement in the US depicts a white Persian cat as aloof and spoilt (Delgado, Munera and Reevy, 2012). Despite the study’s expectations, no significant differences between ratings were seen in black cats (Delgado, Munera and Reevy, 2012). However, this study was conducted in California and so results should not be generalised to the UK, as perceptions may vary between cultures (Delgado, Munera and Reevy, 2012). Furthermore, the study design was solely based on an anonymous questionnaire, which may have led to subject bias, therefore potentially impacting on the results’ reliability (Delgado, Munera and Reevy, 2012). Despite this, Delgado, Munera and Reevy (2012) used a standardised, thus easily repeatable, method using a 7- point Linkert scale, therefore increasing the robustness of the study (Delgado, Munera and Reevy, 2012). Finally, Delgado, Munera and Reevy (2012) focused on how people perceive cat coat colour to affect their personality, and so it varies from the study being proposed as it does not look at how coat colour affects the length of time taken to be adopted (Delgado, Munera and Reevy, 2012). On the contrary to the Delgado, Munera and Reevy (2012) study, a study by Lepper (2002) found that white, colour point and grey cats were most likely to be adopted, whilst black and brown cats were least likely to be adopted, suggesting that coat colour affects adoptability (Lepper, 2002). Therefore, based on their findings, Lepper (2002) considers coat colour as being a factor that impacts on adoptability. In addition to this, Lepper (2002) also found that age and sex of cats were factors that influenced adoptability. Lepper (2002) concluded that the likelihood of adoption decreased as age increased and that spayed male and female cats were more readily adopted than intact males and females, yet intact males were more readily adopted than sexually intact females (Lepper, 2002). Despite this study being similar to the one being proposed, it did not look into the adoptability of ginger or tortoiseshell coloured cats, and these coat colours should also be explored (Lepper, 2002). Furthermore, the study was conducted in California, and so the results may not apply to the UK (Lepper, 2002). This is as adoption preferences may be affected by adoption policies (Lepper, 2002). Therefore, shelters that have adoption policies that greatly differ from Californian shelters may find varying adoption preferences, highlighting the need for future research in areas such as the UK (Lepper, 2002). Furthermore, the study was conducted over a 9- month time frame, omitting the summer period. Therefore, findings may not apply to summer months as the driving factors for adoption, such as preferences, may change over the course of a year or depend on seasons (Lepper, 2002). Therefore, future research is needed throughout the year to control for any preference change (Lepper, 2002).
Contrary to both Delgado, Munera and Reevy (2012) and Lepper (2002), other research has found that coat colour is not a factor that impacts upon adoptability (Fantuzzi et al, 2010). For example, Fantuzzi et al (2010) found that coat colour did not affect the amount of times or duration a cat was viewed in person by potential adopters. They further concluded that coat colour did not affect adoptability as people did not adopt on the basis of this (Fantuzzi et al, 2010). This study differs to the one being proposed, as duration of viewing times was investigated, not the length of stay at the shelter for each coat colour category (Fantuzzi et al, 2010). Furthermore, Sinn (2016) conducted a survey investigating the reasons as to why people adopted their cat (Sinn, 2016). Sinn (2016) found that people primarily adopted a cat for companionship and due to behavioural factors, such as playfulness and willingness to interact with the adopter (Sinn, 2016). Cage location was also found to influence adoptability (Sinn, 2016). Therefore, cat coat colour was not found to be a major factor influencing adoptability (Sinn, 2016). However, the survey only received a response rate of 41%, which included 97 participants from Waterford, Virginia (Sinn, 2016). Despite being an average sample size, a larger sample size would be needed to more fully investigate adopter preference along with participants from a variety of areas as preference may vary between areas and cultures (Sinn, 2016).
On the other hand, a study by Kogan et al (2013) found evidence that black cats take longer to adopt, followed by cats with primarily black coats with other colours mixed in (Kogan et al, 2013). These findings may be due to the negative association with black cats, which are often seen as superstitious (Kogan et al, 2013). However, the cat coat colour categories looked at were black, primarily black with other colours and other colours (Kogan et al, 2013). This grouping method is broad and does not consider variance between coat colours such as brown/ tabby cats, ginger, tortoiseshell and white cats as they are all placed in the ‘other colours’ group (Kogan et al, 2013). Therefore, future research should explore the length of adoption time between distinct coat colour categories, not mixed categories, in order to gather more generaliseable findings and to be able to consider differences between a larger number of groups (Kogan et al, 2013). Despite this, the study used a large sample size, as 29,215 cats were used overall from two different shelters, therefore increasing the studies generalisability (Kogan et al, 2013). Nonetheless, this study states that care must be taken when generalising results to other areas of the US as it was performed in Colorado, and so these findings may not apply to the UK (Kogan et al, 2013).
It is important that the effect of coat colour on cat adoptability is explored in the United Kingdom, as current research is limited and contradictory and would therefore benefit from further research. Potentially, findings may show that a certain coat colour is less favourable to adopters and so those cats remain in shelters for longer, potentially causing the cat to perform stress- induced behaviours that may negatively impact upon their welfare (Kogan et al 2013). In this case, efforts should be made by shelters and organisations who work alongside shelters to promote the coat colour to the public in order to increase their adoptability (Bradshaw et al, 2012). Efforts could include the use of social media to promote coat colour, as information can be aimed at target audiences (Lee and Kotler, 2011). An example of this could be the use of social media pages such as Facebook, as it is a popular platform and so videos promoting different coloured cats could be viewed by many people, potentially increasing shelter cats’ adoption (Lee and Kotler, 2011). In addition to this, potential adopters could be educated prior to adoption about less favourable coat colours and how these cats need to be adopted, such as on the ‘Advice and Welfare’ section of the RSPCA website (RSPCA, 2017). Depending on the study’s findings, staff could state the equal benefits of owning less popularly coloured cats compared to more popularly coloured ones ones, in addition to highlighting the individual personality traits of each cat (RSPCA, 2017).
1.3 Aims and Objectives
The aim of this study is to assess whether cat coat colour affects the length of time taken to be adopted from a rescue shelter. To achieve this aim, the following objectives would need to be met:
1) Analyse the number of days it takes for five different cat coat colours from RSPCA Birmingham Animal Shelter to be adopted over a two-year period (2012-2014). This information will be collated from adoption records held by RSPCA Headquarters to determine the period of time cats spent in the shelter prior to adoption. The colours analysed will be black, black and white, brown/ brown tabby, ginger/ tortoiseshell and white/ mainly white variations (includes colour point). The colour categories chosen are based on the Lepper (2002) and the Kogan et al (2013) study.
2) Determine the current public cat coat colour preference by the use of a questionnaire. The use of a questionnaire will determine if there are any current preferences within the UK population for specific coat colours. The study will also investigate whether these preferences relate with the adoption findings from the first part of the study in order to investigate the relationship, if any, between length of adoption time and public colour preference.
Depending on the outcome of the study, recommendations for marketing of certain coat colours could be made, such as shelters displaying videos of less preferred coloured cats and highlighting why they should be adopted in addition to having regular adoption days devoted to these cats. This recommendation is much like Black Cat Adoption Day although these days could occur on a more regular scale (CatsProtection, 2017). Depending on the findings, the current study may find that there is a coat colour other than black that is least preferred by adopters, thus adoption days could be held for other coloured cats in addition to Black Cat Adoption Day (CatsProtection, 2017)
1.4 Alternative and Null Hypothesis
The alternative hypothesis for this study is that based on the limited amount of research undertaken, black cats and/or black and white cats will take the longest time to be adopted due to the negative stigma that black cats bring bad luck (Kogan et al, 2013). However, it is important to note that findings may differ due to research being so limited and based primarily in the United States (Kogan et al, 2013). The null hypothesis is that there will be no significant difference between the lengths of time that different coat coloured cats spend in shelters.
2.1 Cat Adoption Records: Data Collection
In order to investigate how coat colour affects cats’ length of stay in shelters, retrospective data of cat adoption records from RSPCA Birmingham Animal Shelter were used. Consent was granted by RSPCA Headquarters to use the records they provided. The records used were from the years 2012-2014 and were provided in an Excel document. The use of retrospective data allows data to be gathered from a large time frame, which would not be possible within the given time-frame if using prospective data (Sahai, 1996). RSPCA Birmingham Animal Shelter was chosen due to having a large animal capacity, therefore allowing for a large sample size. A larger sample size allows for reduced anomalous data, therefore increasing the results’ reliability (Hawkins, 2014). This form of data collection is initially qualitative, secondary data, as the information gathered was from archival research (Creswell, 2014). However, this information was then used to calculate a total number of days spent at the shelter, therefore becoming quantitative data as the number of days are a form of numerical data. The data provided by the RSPCA headquarters included: the cat’s Reference Number, Breed, Sex, Colour, Date of Birth, Arrival Date, Age at Arrival, Ready for Rehoming Date, Adoption Date and Age at Departure.
2.1.1 Cat Adoption Records: Ethical Considerations
Each subject cat was identified by its unique reference number rather than a name in order to ensure the cats’ confidentiality by removing personal details (The Data Protection Act, 1998). In addition to this, details of the adopters are not included within the study in order to protect owner confidentiality (The Data Protection Act, 1998).
2.2 Cat Adoption Records: Sample Size
In total, the data of 867 cats consisting of 14 breeds were used (Abyssinian n=2, Bengal n= 3, British Shorthair Cross n= 2, Chinchilla n=1, Domestic Longhair n= 48, Domestic Longhair Cross n=21, Domestic Semi n=18, Domestic Semi Cross n=18, Domestic Shorthair n= 475, Domestic Shorthair Cross n= 272, Exotic Shorthair n=1, Oriental Short n=1, Persian n=1, Persian Cross n=1 and Ragdoll n= 3). The mean age of cats used was 2.5 years old. In order to achieve the objectives of this study, the data analysed included coat colour and time taken to be adopted. However, additional variables were included in the study, such as age, sex and time taken to treat behavioural/ medical issues (arrival date- ready for rehoming date).
The cats used were aged between 1-8 years (1 Year (Approx) n= 271, 2 Years (Approx) n=262, 3 Years (Approx) n=138, 4 Years (Approx) n= 91, 5 Years (Approx) n= 58, 6 Years (Approx) n= 25, 7 Years (Approx) n= 9, 8 Years (Approx) n= 13). Kittens (< 1 year) and elderly cats (+ 8 years) were excluded as they may affect the results. A study by Wiess et al (2012) found that kittens attract adopters’ attention due to their ‘cuteness factor’ and are therefore more popular in comparison to adult cats (Weiss et al, 2012). Based on this study, kittens/ younger cats may potentially not stay in the shelter for as long as older/elderly cats due to their popular appearance, and so including them in the study may affect the length of stay results (Weiss et al, 2012). In addition, data were collected from both male and female cats, as sex may impact adoptability and so this factor needs to be considered (Lepper, 2002). Out of the 867 cats assessed, 460 were female and 407 were male. Sinn (2016) found that adopters primarily adopted their cat due to behavioural factors such as playfulness and willingness to interact (Sinn, 2016). If a cat is not behaviourally sound and is experiencing behavioural issues such as aggression or fearfulness, this may affect adoptability as the cat may not be as playful or willing to interact and may be more withdrawn (Little, 2012). In addition to this, Weiss et al (2012) found that cats’ health and behaviour were important adoption factors (Weiss, 2012). Cats that are in poor health may be adopted less readily (Weiss et al, 2012). Therefore, behavioural and medical factors needed to be controlled for by ensuring each cat used in the study was fit and healthy prior to adoption. Every cat included in the study had been given a period of treatment for issues such as these prior to being made ready for adoption. This treatment period was from the initial arrival date to the ready for rehoming date.
Cat coat colours were categorised based on 5 main groups. The colours assessed were Black (n= 116), Black and White (n=245), Brown/ Brown Tabby (n= 201), Ginger/ Tortoiseshell (n= 186) and White/ Mainly White variations (includes colour point) (n=119). The colour categories chosen were based on the Lepper (2002) and the Kogan et al (2013) study. The groups were designed to be broad due to the inability to consistently code detailed coat colours, which may lead to unreliable data (Kogan et al, 2013). Cats with coat colours that did not fit into these categories such as ‘blue’ and ‘grey’ were removed from the study. Ginger and tortoiseshell colours were placed in the same category, despite many tortoiseshell cats having black in their coat colour, as these colours are reliant on gender. If they were separate categories, gender may have affected these results.
2.3 Cat Adoption Records: Data Analysis
2.3.1 Calculating Time Taken to be Adopted
The time taken to be adopted was calculated by considering the number of days between the Ready for Rehoming date and the Adoption Date for each of the 867 cats. The ready for rehoming date was considered rather than the initial arrival date as processing the data in this way meant that any medical and behavioural issues could be controlled for. As treatment time was not included, as they were not yet ready to be rehomed, these issues should not have impacted upon the results. The time taken per cat to be adopted was measured in days, rather than months as some cats took less than a month to be adopted, and so using days gave a more accurate measure. This data was then used to investigate whether there were significant differences in the time taken to be adopted between coat colour categories.
2.3.2 Cat Adoption Records: Statistical Analysis
The Statistical Package for the Social Sciences (SPSS) was used in order to identify any significant effects of coat colours upon time taken to be adopted. The dependant variable analysed was the time (days) taken to be adopted and the independent variable was coat colour. The Kolgomorov- Smirnov (KS) test was initially performed to determine if the number of days the cats spent at the shelter was normally distributed. The KS test identified that the data was not normally distributed (p = < 0.001) and was therefore non-parametric. As the data were not normally distributed, a Kruskal-Wallis test of difference was performed (Gravetter and Wallnau, 2015). This test was chosen as not only was the data non parametric, but also this study was investigating five unrelated coat colour groups. This test can be used to compare more than three groups, unlike the Mann- Whitney test, which is limited to comparing two groups. (Gravetter and Wallnau, 2015). The Kruskal-Wallis test was performed in order to identify whether there was any significant differences within the data, but does not identify where the significant difference was between the colour groups tested (Gravetter and Wallnau, 2015). Therefore, a post hoc test was needed. The Dunns-Bonferroni approach is a post hoc test which conducts pairwise comparisons on non- parametric data (Hinton, 2014). The purpose of the Dunns-Bonferroni test was to identify where the significant differences in length of stay were found between the coat coloured groups, therefore identifying whether there were any significant differences between the length of time that different coat coloured cats spend in shelters. Brown et al (2013) conducted a similar statistical analysis when comparing how dogs’ age, sex, breed and coat colour influenced their length of stay in shelters (Brown et al, 2013). All statistical tests used were 2 tailed, with the significance level set at p= 0.05, and were run on SPSS version 23 (2015).
2.4 Public Opinion Questionnaire: Design and Sample Size
In addition to the analysis of cat adoption records, a brief, anonymous questionnaire was also conducted in order to investigate the cat coat colour preference of the public within the UK . The questionnaire was created using SurveyMonkey which is an electronic online platform. The questionnaire was uploaded online for a duration of 2 months (November, 2016- January, 2017) and was linked via Facebook. Using an electronic questionnaire has benefits such as the ability to gain a large number of respondents from varying populations and demographic backgrounds. This can allow access to populations which would not be as accessible in person (Wright, 2005). For example, survey monkey allows the questionnaire to be sent over social media sites such as Facebook as was performed for this study. Therefore, when the questionnaire was shared, participants from across the UK were able to access the questionnaire, increasing the response rate and sample size. This increases the validity and accuracy of the data collected (Wright, 2005). In addition to this, providing the questionnaire electronically also increased participant access as they did not have to be sent the questionnaire directly or have it distributed face to face (Wright, 2005). However, controlling for selection bias when using online questionnaires is difficult. There are undoubtedly individuals who are more likely to complete the questionnaire than others and so this must be taken into account (Wright, 2005). Therefore, acquiring a large sample size is necessary as it may help to control for this concern (Wright, 2005).
The questionnaire was designed to gather both qualitative and quantitative data, which was used to gain descriptive statistics. Quantitative questions included selecting a coat colour they least and most preferred, whereas qualitative questions asked why the participants had this preference. The coat colour categories given to the participants to choose from were based on the Lepper (2002) and the Kogan et al (2013) study. Thus, they were the same groups used during analysis of the cat adoption records (Black, Black and white, Brown/Tabby, Ginger/ tortoiseshell and White/ White variations).
The use of a questionnaire helps to identify if there are any current preferences within the UK population for specific coat colours, and whether these potential preferences coincide with the findings of the adoption records part of the study. The sample method used was opportunistic as the questionnaire was available via Facebook, for any participant who fit the criteria to complete (Russel et al, 2004). However, results may not be generalisable back to the target population when using opportunistic sampling (Russel et al, 2004). Therefore, participant criteria were used in order to define the target population (Russel et al, 2004). Participants were made aware in question one that they needed to fit specific criteria in order to be able to complete the questionnaire. The criteria were as follows: must be over the age of 18, can be an owner/non- owner of a cat and must be part of the UK population. Overall, the responses of 99 participants were used, all of whom fit the criteria.
2.4.1 Public Opinion Questionnaire: Ethical Considerations
The questionnaire was designed to be anonymous in order to protect the participants’ data protection rights and to ensure confidentiality (The Data Protection Act, 1998). In addition to this, details such as the participants’ exact location was not asked in order to help maintain participant confidentiality (The Data Protection Act, 1998). Each participant was informed at the beginning of the questionnaire that if they wished to withdraw, they may do so by emailing the address given (The Data Protection Act, 1998). Every participant included in the study gave their consent to be used in the study and ticked the box to state that they were over the age of 18 and fit the criteria (The Data Protection Act, 1998). Participants who did not answer ‘yes’ to question one to give their consent to be used in the study were removed. In total, 106 participants completed the questionnaire. However, seven participants were removed from the study due to skipping this question, thus not giving their consent for their responses to be used.
3.1 Cat Adoption Records Results
3.1.1 Cat Adoption Records: Descriptive Statistics
Descriptive data of the mean number of days taken for each coat colour category was calculated (See figure 1). Black and White cats took the longest time to be adopted (M= 39.5), followed by Black cats (M= 34.1). Brown/ Tabby cats were found neither to spend the longest or shortest duration in RSPCA Birmingham Animal Shelter (M= 29.6), similarly to Ginger/ Tortoiseshell cats (M= 28.1). White/ Mainly White Variation cats spent the shortest mean duration in the shelter (M= 27.8). The overall range of the Time Taken to be adopted was 249 days.
3.1.2 Cat Adoption Records: Statistical Analysis Results
A Kruskal- Wallis test of difference identified that there was a significance difference between cat coat colour categories and the number of days taken to be adopted (x2 (4)= 24.699, n1=116, n2= 245, n3=201, n4=186, n5=119, N= 867, P=<0.001). Therefore, the null hypothesis was rejected. The post hoc Dunns-Bonferroni test was then conducted to identify where the significant difference was between the five coat colour groups. The Dunns- Bonferroni test identified significant differences between the ginger/ tortoiseshell and black and white group (p= <0.001), the white/ white variations and black and white group (p= <0.012) and the brown/tabby and black and white group (p= <0.001). Table 1 shows the significant differences between groups from the Dunns-Bonferroni analysis, and table 2 shows the individual differences between the mean rank scores of these groups. There was a mean rank score of 451.73 for black cats, 495.27 for Black and White cats, 404.62 for Brown/ Tabby cats, 392.87 for Ginger/ Tortoiseshell cats and 404. 49 for White/ Mainly White variation cats
3.2 Public Opinion Questionnaire Results
Of the 99 participants who completed the questionnaire, 82 owned or had previously owned a cat, and 17 respondents had not ever owned a cat. Of these 82 participants, 51 adopted their cats, whilst 16 bought their cat and 15 participants had both bought and adopted a cat.
3.2.1 Coat Colour Most Preferred by Participants
Participants were asked to choose a coat colour that they would most prefer to adopt (figure 2). Out of the 99 participants who completed the questionnaire, 95 responded to this question. Overall, cats that were in the Ginger/Tortoiseshell group were most preferred (34.74%- chosen 33 times), with black cats being the second most preferred (22.11%- chosen 21 times). The Black and white category was chosen 18 times (18.95%) and the Brown category was chosen 16 times (16.84%). White/ mainly white variations were chosen least often (7.37%- chosen 7 times).
When asked to clarify why they chose this category, participants who selected the ginger/tortoiseshell group responded with preferences such as:
- ‘I like the markings’
- ‘It’s very distinctive’
- ‘Seems cuter and softer’
- ‘It’s a nice, warm colour’
- ‘Orange cats always appear to have great personalities’
- ‘I find ginger cats to be friendlier’
- ’Based on the previous cats in our household/colours of cats in films I like’
- ‘I much prefer tabby or tortie colouring, I had a tabby as a child and so have grown up with the colour’
- ‘I have always had ginger or tortoiseshell cats since childhood’
- ‘Just think they have so much more character’
- ‘Ginger looks unique’
- ‘I think that’s the most attractive colour’
- ‘Had a ginger cat in the past’
- ‘Always had a ginger or a tortoiseshell’
3.2.2 Coat Colour Least Preferred by Participants
Participants were also asked to choose a cat colour they would least prefer to adopt (figure 3). Of the 99 participants who completed the questionnaire, 91 responded to this question. White/ mainly white variations were least preferred by the majority of participants (59.34%- chosen 54 times). The brown category was the second least preferred (14.29%- chosen 13 times). The Ginger/ tortoiseshell category and black category were both chosen 10 times and so were equally preferred (10.99%). Lastly, the black and white category was chosen 4 times (4.40%). However, there was little variation between the amount of times that the brown, ginger/tortoiseshell, black and black and white categories were chosen in comparison with white/ mainly white variations.
When asked to clarify why participants least preferred the colour chosen, respondents who chose the white/ mainly white variations category gave responses such as:
- ‘Never had one, white is a boring colour, therefore I associate with a potentially boring character’
- ‘More prone to skin cancer and sunburn’
- ‘Plain- white short haired cats aren’t my type’
- ‘Sterile, not welcoming’
- ‘Not so appealing’
- ‘Often susceptible to cancer of ears and nose’
- ‘I think that a white cat would look grubby after a while’
- ‘Cat fur on clothes’
- ‘Risk of skin cancers’
- ‘Historical, the first cat I disliked/felt scared of was white. Also, white cats tend to be long-haired, with all the problems that could bring’
- ‘They usually look grubby and their coat doesn’t look as nice’
- ‘May look dirty’
- ‘I believe pure white cats are often deaf’
- ‘Associated health problems’
4.1 Findings from the Adoption Records Analysis
The main aim of this study was to determine whether cat coat colour affects cat adoptability. Results from the analysis of the adoption records show that there is a statistically significant effect of cat coat colour upon the number of days taken to be adopted, therefore suggesting that the colour of a cat impacts on the time taken for the cat to be adopted from a rescue shelter. These findings conflict with previous studies that have found that coat colour does not impact adoptability (Sinn, 2016; Fantuzzi et al, 2010). Differences between findings may have been due to the current study being conducted in the United Kingdom, whereas contrasting research was conducted within the United States (Sinn, 2016; Fantuzzi et al, 2010; Lepper, 2002). Thus, findings may differ due to contrasting cultures and shelter adoption policies (Sinn, 2016; Fantuzzi et al, 2010; Lepper, 2002).
When means were considered, black and White cats were found to take the longest time to be adopted, followed closely by black cats. Thus, the alternative hypothesis was accepted. The main significant differences in the length of adoption time were between the ginger/ tortoiseshell group and the black and white group, the white/ white variations and black and white group and the brown/tabby and the black and white group. These results indicate that the number of days that black and white cats take to be adopted differ significantly in comparison to ginger/ tortoiseshell cats, white/white variation cats and brown/tabby cats. It is interesting to note that no significant differences were found between the black and black and white group and the other groups compared to each other. The white/ white variations group took the least amount of time to be adopted, which were closely followed by ginger/ tortoiseshell cats. These findings support those found by Lepper (2002), who concluded that white, colour point and grey cats were most likely to be adopted (Lepper, 2002).
In other research, black cats have been found to take the longest amount of time to be adopted due to superstitious beliefs that black cats bring bad luck (Kogan et al, 2013). Unusually, in this study, this was not found. Both Lepper (2002) and Kogan et al (2013) found that black or brown cats took the longest time to be adopted. Lepper (2002) found strong evidence to suggest that black and brown cats take the longest time to be adopted, whilst Kogan et al (2013) found that black cats take the longest time to be adopted. Neither of these studies found similar adoption concerns for black and white cats (Kogan et al, 2013; Lepper, 2002). The differences in findings between this study and the Kogan et al (2013) study may be due to how coat colours are grouped. For example, the groups used in the Kogan et al (2013) study were black, primarily black with other colours and other colours (Kogan et al, 2013). This grouping method is broad and does not consider colours such as black and white, tortoiseshell and tabby (Kogan et al, 2013). Therefore, individual differences between these colours would not be seen as they would have been broadly placed into groups. This could potentially explain why black and white cats were not found to take the longest time to be adopted (Kogan et al, 2013). In addition to this, the grouping method in the Kogan et al (2013) study allows for subjective interpretation as many different coloured cats fall into varying groups and so differentiating which cat should go into which group becomes challenging (Kogan et al, 2013). Due to this, the primarily black group would be the easiest group to interpret and subjective interpretation of a large range of other colours may have impacted on the other groups (Kogan et al, 2013).
Differences between this study and the Lepper (2002) and Kogan et al (2013) studies may have been due to the sample size of the varying groups within the study. The black and white group had the largest sample size (n=245) in comparison to the other four groups examined (black n=116, brown/tabby n=201, ginger/ tortoiseshell n=186, white/ mainly white variations n=119). Results may have therefore been affected by the differences in sample sizes within groups, which in turn may cause biased results. However, this factor would have been difficult to control for, as these were the real sample sizes of cats adopted from RSPCA Birmingham Animal Shelter during the years 2012- 2014. Removing cats from the sample in order to generate even group sample sizes would have impacted on the results’ reliability and led to potential bias as the sample would no longer be relevant to the real world context. In addition, these sample sizes indicate that over the two- year time period, the majority of cats aged 1-8 that RSPCA Birmingham Animal Shelter received were black and white cats. Considering this information alongside the results of this study highlights the urgency of getting black and white cats rehomed faster.
Reasons as to why black and white cats take longer to be adopted could be purely based on colour, as factors such as age, behavioural/medical issues and sex of the cats were controlled for. In addition to this, previous research has identified that darker coloured cats are less popular due to superstitious beliefs and so this may be the same for black and white cats (Kogan et al, 2013). It is also interesting to note that due to black and white cats being more prevalent within the shelter during the time frame (2012-2014), adopters may have chosen them less readily as there were less of the other coloured cats housed in the shelter. Thus more interest may have been shown in the other coloured cats who were not black and white. This may potentially explain why black and white cats took longer to be adopted. Despite this, there are a multitude of factors that may affect cat adoptability. For example, studies have found that cats housed at eye level with a greater number of toys were adopted more readily (Fantuzzi et al, 2010). Additionally, research has found that cage height impacts adoptability (Sinn, 2016). Factors such as these were not controlled for in this study as they were not included within the data-set, and so it is not known how much impact such as cage height and level of enrichment had upon adoptability (Sinn, 2016; Fantuzzi et al, 2010). Despite this, it is important that rescue shelters consider the findings of this study, as this was one of the first studies investigating the impact that coat colour has on adoption within the UK. Further research could look into other factors that affect adoption within the UK. In addition to coat colour, factors such as cage height and level of enrichment could be explored. It is recommended that shelters still have a day where black cat adoption is promoted, however, it is vital that black and white cats are included in this promotion as it is clear from this study that black and white cats, not only come into shelters in large numbers, but also take a significantly longer time to be adopted. Since these cats could be experiencing chronic stress, reducing the time taken for these animals to be adopted is imperative (Rodan and Heath, 2016). However, this data was only conducted using one rescue shelter within the UK, and so further research is needed using a larger number of shelters in order to increase the reliability and generalisability of results.
Interestingly, the questionnaire did not find that black and white cats were least preferred, instead white and white variation cats were least preferred. This contradicts the findings of the earlier part of the study where as white/ mainly white variation cats were found to take the least time to be adopted. However, it was found that ginger and tortoiseshell cats are most preferred by respondents which corresponds to the adoption records analysis, as these cats were adopted readily in comparison to both black and black and white cats. Differences between the adoption records analysis and questionnaire may be due to the small sample size of the questionnaire and the fact that data was only collected from one shelter. The finding that black and white cats take longer to be adopted may be dependent on the shelter sample and time frame. Therefore, further research is needed using a larger number of shelters in order to increase the reliability and generalisability of the results. Despite this, the findings from this study should still be considered and efforts to increase black and white cat adoption should be implemented. Efforts could include animal shelters holding a black and white cat adoption day per month, alongside featured videos promoting black and white cats on organisations’ social media sites, such as Facebook (Lee and Kotler, 2011).
4.2 Preferences for Cat Coat Colour According to the UK Public
Results from the questionnaire revealed that ginger and tortoiseshell coloured cats were the most preferred, whereas white and mainly white variation cats were the least preferred. These findings conflict with the results from the adoption records analysis, where white/ white variation cats took the least amount of time to be adopted. Despite this, respondent preferences from the questionnaire are supported by the Delgado, Munera and Reevy (2012) study. Delgado, Munera and Reevy (2012) conducted a survey to investigate whether people may perceive cat coat colour as a factor that contributes to the personality of cats. Delgado, Munera and Reevy (2012) discovered that orange cats were rated relatively highly in friendliness due to people associating ginger cats with the popular TV show Garfield, whereas white cats were rated as shy, more aloof and calm. These findings may suggest why participants in this study preferred ginger/ tortoiseshell cats and least preferred white/ white variation cats (Delgado, Munera and Reevy, 2012). Preferences from the questionnaire in the current study coincide with the findings of Delgado, Munera and Reevy (2012), as respondents stated their reasons for most preferring ginger/ tortoiseshell cats were such as ‘I find ginger cats to be friendlier’, ‘Orange cats always appear to have great personalities’, ‘Just think they have so much more character’ and ‘colours of cats in films I like’. In addition to this, respondents who least preferred white/ white variation cats gave reasons such as the colour being ‘sterile, not welcoming’ and ‘Never had one, white is a boring colour, therefore I associate with a potentially boring character’. Therefore, people may perceive coat colour as a factor that contributes to the personality of cats (Delgado, Munera and Reevy, 2012). In addition, Gourkow and Fraser (2006) concluded that the majority of adopters that were involved in their questionnaire selected a cat for companionship in addition to behavioural and emotional traits of the cat. Level of feline affection has been positively associated with levels of playfulness and curiosity, in addition to general cleanliness of the cat (Turner and Stammbach-Geering, 1990). This coincides with the findings of Sinn (2016), who found that adopters primarily adopted a cat for companionship and factors such as playfulness and willingness to interact with the adopter (Sinn, 2016). Therefore, people may choose a cat based on its colour due to preconceptions that coat colours associate with different behavioural traits, and when looking for an ideal trait (such as playfulness and curiosity), people may not choose a particular cat colour as they believe that this does not show the desired behavioural trait (Delgado, Munera and Reevy, 2012; Gourkow and Fraser, 2006). Relationships between coat colour and behaviour have been suggested in previous research (Turner and Bateson, 2000). For example, the lack of pigment in the iris of albino animals has been suggested to cause problems with visual perceptions, thus causing underlying behavioural characteristics such as sluggishness of reactions (Turner and Bateson, 2000). Another possibility is that genes coding for coat colour are located at positions on chromosomes close to other genes that have a partial influence on the function of the nervous system (Turner and Bateson, 2000). For example, blue eyed white cats are often deaf (Turner and Bateson, 2000). This genetically induced defect has a marked effect on behaviour and has thus caused breeders to regard these animals as ‘dull of intellect and slow in thinking’ (Turner and Bateson, 2000). White coat colour may therefore be positioned close to a gene which induces deafness. Thus cats that are inheriting one trait are likely to inherit the other (Turner and Bateson, 2000). People may perceive white cats as unable to be playful and curious due to preconceptions that white cats are often dull and slow (Turner and Bateson, 2000). In addition to this, many participants stated that they least preferred white cats due to associated health risks such as deafness and risk of skin cancers. People also least preferred white/ white variation cats due to white cat fur showing on clothes more easily, and white cats looking ‘grubby’. This was stated numerous times. Participants may least prefer cats that look ‘grubby’ as the level of affection with a cat has been positively correlated with the general cleanliness of the cat in addition to playfulness and curiosity (Turner and Stammbach-Geering, 1990). Therefore, participants may have felt that they would bond less with a white or mainly white cat as they perceive these cats as being generally less clean (Turner and Stammbach-Geering, 1990).
It is interesting to note that these preferences may be in relation to purely white cats, as it is the purely white colour that causes associated health risks, such as deafness (Turner and Bateson, 2000). Therefore, respondents potentially viewed this group as purely white cats rather than white and mainly white variations. Respondents may have not known that colour point cats would in fact fall into this group, which commonly have brown facial features but a white body. Previous research by Lepper (2002) found that colour point cats are more likely to be adopted. The inclusion of colour point cats within the white/ mainly white group may therefore be why this group was found to take the least time to be adopted in the adoption records analysis in comparison to black, black and white, brown/ brown tabby and ginger/ tortoiseshell cats, as colour point cats have been found to be popular in previous research (Lepper, 2002). Therefore, due to participants potentially not perceiving colour point cats as being within the white/ mainly white variation group, respondents may have based their preference on solely white cats. This may potentially be why white/ white variation cats were found to be least preferred by respondents and therefore why conflicting results were found between the adoption records analysis and the questionnaire. Future research would benefit from splitting the white/ white variation groups into two separate groups to avoid confusion. Additionally, the questionnaire only gathered 99 participants and so differences between the statistical analysis and questionnaire are based on a small sample. A larger sample size would be needed to fully investigate this topic.
It was also found from the questionnaire that early childhood experiences may form cat coat colour preferences (Pressley and McCormick, 2007). For example, participants who most preferred ginger/tortoiseshell cats gave reasons such as ‘I had these as a child and so have grown up with the colour’ and ‘I have always had ginger or tortoiseshell cats since childhood’. Additionally, participants who least preferred white/ white variation cats gave reasons such as ‘the first cat I disliked/felt scared of was white’. Therefore, early experience of an animal may influence later opinions on other animals. This is supported by Watson and Rayner (1920) who found that early childhood experiences can form aversive responses and phobias towards animals later in life (Watson and Rayner, 1920). For example, Watson and Rayner (1920) classically conditioned the nine- month old ‘Little Albert’ to become afraid of white animals. The onset of classical conditioning caused ‘Little Albert’ to form aversive feelings towards the white animals and therefore led him to become phobic of them later on in life (Watson and Rayner, 1920). Thus, it is important that organisations promote less favourably coloured cats to potential adopters in order to avoid aversion of certain colours. For example, the individual personality of the cat should be highlighted to adopters in order to avoid adoption being based solely on preconceptions and early experience of colour (RSPCA, 2017). However, since these findings are based on participant responses to open-questions, which is a form of qualitative data, this is subject to bias. Therefore, future research would benefit from gathering quantitative data investigating the direct effects of early childhood experiences on the adoption of cats in order to identify if this is a factor that significantly impacts cat adoption.
Based on these findings, people may perceive that certain coat colours account for the personality of the animal. Thus, people may ignore cats with certain coat colours due to preconceptions that some colours are more or less likely to display desired behavioural traits, such as playfulness and curiosity (Sinn, 2016; Turner and Stammbach-Geering, 1990). However, further research exploring this is needed. In addition to this, people also perceived coat colours as being associated with health risks, and so are less likely to adopt animals if they perceive that the colour is more prone to develop health problems. This is supported by Weiss (2012), who found evidence that cats’ health and behaviour were important adoption factors (Weiss, 2012). Early childhood experience or prior experience could also influence both positive and negative preferences for cat colour, thus impacting adoptability of cats. Findings from the questionnaire can therefore be used to determine why people have preferences and how these preferences directly relate to the colour of a cat.
4.3 Limitations and Future Research
Only one shelter (RSPCA Birmingham Animal Shelter) was used in the study. Therefore, the generalisability of the results is impacted as it is not known if these adoption trends are shown by other shelters. Despite this, the adoption records used were from a large, 2-year time frame, which were during recent years (2012-2014), thus increasing the current study’s robustness. In addition to this, a large sample size of cats was used in order to increase generalisability, despite sample size varying between groups. However, adoption trends may vary due to the location of the shelter, thus, future research would benefit from using a multitude of shelters from around the UK, in addition to shelters from other organisations such as BlueCross. In addition, coat colour categories could be covered in more depth as there may have been confusion in the questionnaire between what white and white variation colours are. This therefore allowed for subjective consideration by participants, which may have caused the results to become biased. Coat colour categories were attempted to be made more defined, as previous research used even broader groups (Kogan et al, 2013), however, white and white variation cats should be grouped separately in order to avoid confusion for future research. Additionally, participants were not provided with a ‘don’t mind’ option within the questionnaire. Many participants stated that they did not in fact have a preference for colour but were forced to choose one anyway as there was not an option to not choose. This therefore resulted in participants making false choices, which may have impacted the validity of the results as some preferences were forced. Furthermore, this resulted in participants skipping questions, leading to partial completion of the questionnaire. A larger sample size, and greater consideration of approach to questioning, is needed to alleviate these issues. Lastly, questionnaire findings were based purely on consideration of descriptive data. In order to make the findings more quantitative and more robust, statistical analysis could be performed with the use of Likert scales. Likert scales allow for quantitative data to be gathered as participants select their answer based on a scale, which can then be tested for significant differences (Brace, 2008). Although Likert scales are straight forward and quantifiable, they do not allow for participants to express their opinions within open questions, which was required in this questionnaire in order to gather participant preference (Brace, 2008). Despite this, the questionnaire allowed for a general gauge of why people have preferences on coat colour to be gathered, which can be used to help future adoption of cats.
Cat coat colour was found to have a significant effect on time taken to be adopted. Results indicated that black and white cats take the longest time to be adopted in comparison to black cats, brown/tabby cats, ginger and tortoiseshell cats, and white and white variation cats, whereas white/ mainly white variation cats were found to take the least time to be adopted. However, further research is needed to investigate this using data from a number of different shelters around the UK. Results from the questionnaire showed that ginger/ tortoiseshell cats are most preferred whereas white/ white variation cats are least preferred. Conflicting results between the adoption records analysis and questionnaire may have been due to the method of grouping, which may have allowed for misinterpretation of white/ mainly white variation cats. A larger sample sizes is also required to investigate whether findings from the questionnaire are representative In addition, results from the questionnaire revealed some reasons as to why people have preferences in relation to cat colour and adoption. Preferences were found to be based on preconceived information that colour directly relates to behaviour, and so when choosing a cat based on desirable behavioural traits, some colours may be dismissed if the person does not think that the cat colour will show this trait. In addition to this, people have preferences on colour due to believing that certain colours are more prone to associated health issues. Additionally, colours that are seen as looking more dirty may impact on colour preferences, as a lower level of cleanliness is negatively correlated with level of attachment. Furthermore, early childhood experience associated with a certain cat coat colour could impact adoptability both positively and negatively depending on the prior experience.
Based on the findings from this study, it is recommended that animal shelters such as RSPCA use these preferences when promoting coloured cats. For example, the individual personality of the cat should be highlighted to adopters in order to avoid adoption being based purely on preconceptions and early experience of colour. This is similar to how black cats may not be favoured due to superstition and white cats may be deemed as less playful and curious. Furthermore, animal shelters, such as the RSPCA, should be proactive in promoting black and white cats to the public as these cats have been found to take longer to be adopted. Increased adoption rates will decrease the amount of time these coloured cats spend within shelter environments, thus aiming to decrease the occurrence of chronic stress. Efforts to increase black and white cat adoption could include animal shelters holding a black and white cat adoption day per month, alongside featured videos promoting black and white cats on organisation social media sites, such as Facebook. Further research is needed to increase the generalisability of results from both the questionnaire and adoption records analysis. In order to do this, data should be collected from a multitude of shelters from around the UK, in addition to increasing the sample size of the questionnaire. Furthermore, future research would benefit from splitting the white/ white variation groups into separate groups in order to reduce any misinterpretation of colours. Additionally, participants should be provided with a choice within the questionnaire to not choose a colour in order to avoid participants falsely choosing preferences or skipping questions.
Brace, I. (2008) Questionnaire Design: How to plan, structure and write a survey material for effective market research. 2nd ed. London, United Kingdom: Kogan Page.
Bradshaw, J. W. S., Casey, R. A. and Brown, S. L. (2012) The behaviour of the domestic cat. 2nd ed. Oxfordshire, UK: CABI.
Broom, D. M. (1991) Animal welfare: concepts and measurements. Journal of Animal Science. 69 (10), pp.4167- 4175.
Brown, W. P., Davidson, J. P. and Zuefle, M. E. (2013). Effects of phenotypic characteristics on the length of stay of dogs at two no kill animal shelters. Journal of Applied Animal Welfare Science. 16 (1), pp.2-18.
Carlstead, K., Brown, J.L. and Strawn, W. (1993) Behavioural and physiological correlates of stress in laboratory cats. Applied Animal Behaviour Science, 38 (2), pp. 143–158.
CATSPROTECTION (2017) National black cat day 2016. Available at: http://www.cats.org.uk/black-cats [Accessed: 27th February 2017].
Creswell, J.W. (2014) Research design: Qualitative, quantitative, and mixed methods approaches. 4th ed. Thousand Oaks, California: SAGE Publications, Inc.
Delgado M. M., Munera, J. D. and Reevy, G. M. (2012) Human Perceptions of Coat Colour as an Indicator of Domestic Cat Personality. Anthrozoos. 25 (4), pp. 427- 440.
Fantuzzi, J. M., Miller, K. A. and Weiss, E. (2010) Factors relevant to adoption of cats in an animal shelter. Journal of Applied Animal Welfare Science. 13 (2), pp.174-179.
Gourkow, N. (2001) Factors affecting the welfare and adoption rate of cats in an animal shelter. The University of British Columbia, 1, pp. 1–59.
Gourkow, N. and Fraser, D. (2006) The effect of housing and handling practices on the welfare, behaviour and selection of domestic cats (Felis sylvestris catus) by adopters in an animal shelter. Animal Welfare, 15, pp. 371–377.
Gravetter, F.J. and Wallnau, L.B. (2015) Statistics for the behavioral sciences. 10th ed. Boston, MA, United States: Wadsworth Cengage Learning.
Hawkins, D. (2014) Biomeasurement: A student’s guide to biological statistics. 3rd ed. Oxford: Oxford University Press.
Hinton, P.R. (2014) Statistics explained. 3rd ed. London: Routledge.
Holmes, R. J. (1997). Boredom by dogs and cats. Proceedings from the 1997 Federation of Asian Veterinarian Association Congress. pp. 26-29.
Holmes, R. J. (1993). Separation anxiety and grieving by dogs and cats. In University of Sydney, Animal behaviour proceedings 214. pp. 61-64.
Kogan, L. R., Schoenfeld- Tacher, R. and Hellyer, P. W. (2013) Cats in animal shelters: Exploring the common perception that black cats take longer to adopt. The Open Veterinary Science Journal. (7), pp.18-22.
Kry, K. and Casey, R. (2007) The effect of hiding enrichment on the stress levels and behaviour of domestic cats (Felis sylvestis catus) in a shelter setting and the implications for adoption potential. Animal Welfare. (16), pp.375- 383
Lee, N. R. and Kotler, P. (2011) Social Marketing: Influencing Behaviours for Good. United States of America: SAGE Publications.
Lepper, M., Kass, P. H. and Hart, L. A. (2002) Prediction of adoption versus euthanasia among dogs and cats in a California animal shelter. Journal of Applied Animal Welfare Science. 5 (1), pp.29-42.
Little, S.D.E. (2012) The cat: Clinical medicine and management. Philadelphia, PA, United States: Elsevier Saunders.
McCobb, E.C., Patronek, G.J., Marder, A., Dinnage, J.D., Stone, M.S. (2005) Assessment of stress levels among cats in four animal shelters. J. Am. Vet. Med. Assoc. 226, pp.548–555.
Morton, D. B. (1998) Fear. In M. Bekoff: Encyclopaedia of animal rights and animal welfare. Westport, Connecticut: Greenwood Press.
Pedersen, N.C., Pratt, P.W. (1991) Feline Husbandry: Diseases and Management in the Multiple-cat Environment. American Veterinary Publications, Goleta, CA.
Pressley, M. and McCormick, C. B. (2007). Child and adolescent development for educators. New York: The Guilford Press.
Rochlitz, I. (2007). The Welfare of Cats. Dordrecht, The Netherlands: Springer.
Rodan, I. and Heath, S. (2016) Feline Behavioural Health and Welfare. St Louis, Missouri: Elsevier.
RSPCA (2017). Advice and Welfare. Available from: https://www.rspca.org.uk/adviceandwelfare [Accessed on 10th February 2017]
RSPCA (2016). Available from: http://media.rspca.org.uk/media/facts [Accessed on 16th February 2016]
Russell, J., Jarvis, M., Gorman, P., Flanagan, C. and Dolan, L. (2004) Angles on psychology (angles on psychology). 2nd ed. Cheltenham, U.K.: Nelson Thornes.
Sahai, H. (1996) Statistics in Epidemiology: Methods, Techniques and Applications. Florida: CRC Press LLC.
Sinn, L. (2016) Factors affecting the selection of cats by adopters. Journal of Veterinary Behavior: Clinical Applications and Research, 14, pp. 5–9.
SurveyMonkey (2017) Billable- Responses. Available at: https://help.surveymonkey.com/articles/en_US/kb/Billable-Responses [Accessed: 1st March, 2017]
The Data Protection Act (1998). Available from: https://www.gov.uk/data-protection/the-data-protection-act [Accessed: 3rd January 2017]
Turner, B. C and Bateson, P. (2014) The domestic cat: The biology of its behaviour. 3rd ed. Cambridge: Cambridge University Press.
Turner, D.C. and Bateson, P. (2000) The domestic cat: The biology of its behaviour. 2nd ed. Cambridge, UK: Cambridge University Press.
Turner, D. C., Stammbach-Geering, K. (1990) Owner assessment and the ethology of human–cat relationship: Pets, Benefits and Practice. Proceedings of Waltham Symposium 20. London: BVA Publications.
Watson, J.B. and Rayner, R. (1920) Conditioned emotional reactions. Journal of Experimental Psychology, 3 (1), pp.1–14.
Webster, J.G. (2005) Animal welfare: Limping towards Eden; a practical approach to redressing the problem of our dominion over the animals. 2nd edn. Ames, IA: Wiley-Blackwell.
Weiss, E., Miller, K., Mohan- Gibbons, H. and Vela, C. (2012). Why did you choose this pet?: Adopters and pet selection preferences in five animal shelters in the United States. Animals. 2 (2), pp.144-159.
Wright, K.B. (2005) Researching Internet-Based populations: Advantages and disadvantages of online survey research, online questionnaire authoring software packages, and web survey services. Journal of Computer-Mediated Communication, 10 (3), pp. 00–00.