The impact of body diversity vs thin-idealistic media messaging on health outcomes: an experimental study

ABSTRACT

The recent rise in body dissatisfaction and weight bias has led to a call to the media to increase the diversity of their imagery, in efforts to challenge the thin-ideal. Therefore, this study aimed to evaluate the effects of both body diversity and thin-ideal interventions on health outcomes. Female participants (n = 160) were randomly allocated into an intervention group: body diversity; thin-ideal; control. They completed measures of body satisfaction, body compassion, internalisation of the thin-ideal, weight bias and behavioural intentions at baseline and post-intervention. The results showed significant differences between groups for weight bias and intentions to eat healthily. Specifically, those in the body diversity intervention group reported a greater reduction in weight bias compared to the other conditions. Further, those in the thin-ideal intervention group reported a greater increase in intentions to eat healthily compared to the other conditions. There were no differences between groups for body satisfaction, body compassion, internalisation of the thin-ideal and behavioural intentions to exercise and manage weight. In conclusion, exposure to body diversity images reduced weight bias whereas exposure to the thin-ideal promoted intentions towards healthy eating. These findings therefore offer empirical evidence for the impact of using different types of imagery to change different health outcomes.

KEYWORDS:

• Body diversity
• thin-ideal
• intervention
• weight bias
• media

Introduction

Whilst there are many possible causes for the widespread problems regarding body size, eating behaviour and body dissatisfaction, sociocultural theory provides a theoretical framework for understanding the relationships between normative ideals, body dissatisfaction and disordered eating behaviours (Thompson et al., 1999). Current beauty standards are being increasingly challenged, however still favour the thin-ideal, typifying a slender, feminine physique with a small waist and little body fat. Internalising these appearance ideals as standards to meet lead to upward body comparison processes. Body dissatisfaction occurs when there is a perceived incongruence between ideals and their appearance (Thompson et al., 1999). It is also suggested that pressure to conform to these ideals leads to the internalisation of anti-fat and stigmatising attitudes towards those with overweight and obesity through implicit messaging that thinness equates to happiness, success and gain (Rodgers, 2016).

These pressures come from a variety of sources such as friends and family (Quiles Marcos et al., 2013); however, thin-idealistic messaging from the media has attracted most research attention in recent times (Schaefer et al., 2015). A growing body of research illustrates the significant impact of both acute and continual exposure to unrealistic media images and messaging on numerous outcomes such as negative mood, body dissatisfaction, appearance anxiety, maladaptive eating behaviour and physical activity (Bury et al., 2016; Derenne & Beresin, 2018; Halliwell et al., 2005; Mayer-Brown et al., 2016; Moreno-Domínguez et al., 2019; Pedersen et al., 2018; Robinson et al., 2017; Slevec & Tiggemann, 2011; Tiggemann & Miller, 2010).

The relationship between thin-idealistic messaging and body dissatisfaction and negative mood is prevalent across various media domains including television (Anschutz et al., 2011), traditional print media (Tiggemann & McGill, 2004) and most recently, digital media. The ability to share photos and information immediately via social media presents with additional stressors to everyday life. Research has investigated the role of social media exposure on body image in a sample of Australian high school adolescent girls (Tiggemann & Miller, 2010). Accessing websites and reading magazines that place an emphasis on physical appearance correlated with a greater drive for thinness, thin-ideal internalisation, appearance comparison and weight dissatisfaction. More recent research has also considered the role of social media and body satisfaction. For example, a cross-sectional study of adolescent girls found that frequency of social media usage was associated with greater internalisation of beauty ideals (Vandenbosch & Eggermont, 2012).

Interventions have been designed to mitigate the potentially damaging impacts of the media on wellbeing. One approach to reduce the negative impact of the thin-ideal is through societal level legislative interventions such as banning models who are visibly underweight, and including warning labels for digitally altered images; however, the evidence of the efficacy of this is weak (McComb & Mills, 2020). Conversely, research has suggested that higher media literacy, characterised by the ability to think critically about the media in general (Silverblatt et al., 2014), and in particular, being able to evaluate how realistic an image is, protects against the negative influences of the media (Berel & Irving, 1998; Yamamiya et al., 2005).

An alternative approach is to change the types of imagery used in the media, so that they better reflect a range of realistic body types, moving towards a focus on functionality, rather than aesthetic qualities (Mulgrew & Tiggemann, 2018). This has largely been demonstrated with regards to the impact of media imagery on eating and exercise intentions (Alleva et al., 2017; Mulgrew et al., 2018; Robinson et al., 2017). Body positivity is a growing social media trend, challenging conventional attractiveness norms by increasing visibility of a diverse range of body sizes (Cohen et al., 2019). This focus on increased visibility is reflected in weight normalisation research demonstrating that repeated exposure to larger body sizes can lead to a shift in perceptions of body size preferences, and the tolerance and acceptance of obesity (Boothroyd et al., 2012; Robinson & Kirkham, 2014). Media campaigns aiming to promote diversity and self-confidence in women such as ‘This Girl Can’ and ‘#jointhemovement’ have been found to be successful in improving functionality satisfaction and exercise intentions compared with a control video (Mulgrew et al., 2018). Furthermore, the ‘health at every size’ movement aims to promote well-being and body positivity through the eradication of weight bias (Bacon, 2010; Penney & Kirk, 2015; Tomiyama, 2014).

Research has suggested that self-compassion interventions are effective in reducing body dissatisfaction (Albertson et al., 2015), and that media exposure impacts body compassion (Beadle, 2019). Ogden et al. (2020) investigated whether viewing body diversity images, thin-ideal images or household objects (control) could buffer against thin-ideal norms. Findings demonstrated that viewing body diversity images resulted in higher body compassion, body satisfaction, and stronger negative attitudes towards the thin-ideal. Therefore, it is plausible to suggest that interventions that change the sorts of images used in media could not only improve body dissatisfaction, but also reduce weight bias.

To summarise, the media plays a pivotal role in fostering the development of body dissatisfaction through thin-idealistic imagery. The negative impact of thin-idealistic media messaging can be damaging tothose across the weight-span; however, in light of the rise in global obesity prevalence, efforts need to be made to ensure a body diversity approach is taken to increase body satisfaction, health behaviours and decrease weight bias. Interventions to achieve this include approaches to improve media criticality and warning labels. Research has suggested the efficacy of changing the types of media images used to promote body diversity (Diedrichs & Lee, 2011; Halliwell et al., 2005; Hendrickse et al., 2020; Ogden et al., 2020); however, this impact of this on weight bias needs to be further explored.

The present study therefore aimed to evaluate the impact of exposure to body diversity and thin-ideal intervention videos on health outcomes; body satisfaction, body compassion, internalisation of the thin-ideal, weight bias and behavioural intentions to eat more healthily, exercise more and successfully manage weight. Due to the protective role of media literacy against the negative impact of thin-idealistic media messaging (Berel & Irving, 1998; Yamamiya et al., 2005), media literacy was included as a covariate in subsequent analyses. It was hypothesised that exposure to the body diversity intervention video would increase body satisfaction and body compassion, and decrease behavioural intentions, the internalisation of the thin-ideal and weight bias; and that exposure to the thin-ideal intervention would decrease body satisfaction and body compassion, and increase behavioural intentions, the internalisation of the thin-ideal and weight bias.

Materials and methods

Design

The study used an experimental study design with participants being allocated to one of three conditions: body diversity; the thin ideal; control. The independent variables were therefore intervention condition (body diversity, thin-ideal and control), and time (baseline vs. post-intervention). The dependent variables were body satisfaction, body compassion, internalisation of the thin-ideal, weight bias, and behavioural intentions (healthy eating, exercise and weight management). Media literacy was a covariate.

Participants

One hundred and sixty female participants from the general population were recruited from a social media advertisement for a research study investigating women’s beliefs about their bodies. The mean age of participants was 28.2 years (SD = 10.8, range = 16–74). The majority of the sample was white (n = 137, 85.6%), 9 were Asian (5.6%), 3 were Black (1.9%) and 11 classified themselves as other (6.9%). Participants also provided their self-reported BMI classification; 6 participants reported being underweight (3.8%), 118 reported being of healthy weight (73.8%), 30 reported being overweight (18.8%) and 6 reported being obese (3.8%).

The intervention

The authors selected all intervention videos from the internet. The body diversity and thin-ideal videos were commercially developed videos, so consideration was taken to ensure these were reflective of construct definitions in the literature. The body diversity and thin-ideal videos were matched on similar exercise levels and having an upbeat tempo. Participants were randomly allocated into one of the following conditions:

Body diversity

Participants viewed a 1-minute body diversity ‘This Girl Can – Me Again’ campaign video aimed at normalising ‘normal’. The video depicts women with a range of body shapes and sizes exercising, with the message that there is no ‘right’ way to be active, and therefore is consistent with key concepts in the body diversity literature (Cohen et al., 2019; Mulgrew et al., 2018); retrieved from (https://www.youtube.com/watch?v=4BKwk8q4H0Y).

The thin-ideal

Participants viewed a 1-minute video advertising women’s activewear. The video contains thin, conventionally attractive women exercising, conforming to western ideals of beauty; retrieved from https://www.youtube.com/watch?v=mec56sJA3ew. The authors viewed several similar video clips; however, the joint decision was made that this video better reflected the definitions of the thin-ideal in the literature (Mazur, 1986; Schaefer et al., 2015; Thompson et al., 1999; Wiseman et al., 1992).

Control

This 1-minute nature video showed images of plants and the weather accompanied by relaxing meditation music, retrieved from https://www.youtube.com/watch?v=vYPYgHwf2KAandt=1s.

Measures

Demographics

Participants reported their gender, age, self-reported BMI group and their ethnicity.

Media literacy

Participants’ attitudes towards the media were measured through the 6-item critical thinking about media messages scale (CTMM; Scull et al., 2010), ranging from 1 (never) to 5 (always). Items measure the degree to which people report that they think critically about the messages they receive through the media. For example, ‘I think about what the people who made a media message want me to believe’ (α = 0.88).

The following scales were completed before and after the intervention. All time frames were adjusted for ‘how do you feel right now’ to enable change from pre- to post-intervention:

Body compassion

The shortened 16-item version of the Body Compassion and Criticism Scale (BoCCS; Beadle, 2019) was used. It is comprised of four subscales rated on a 5-point scale from 1 (not at all) to 5 (very much): body kindness (4 items: e.g. ‘I accept the flaws in my body, even if I don’t like them’), common humanity (4 items: e.g. ‘I am sure everyone has insecurities about their bodies’), motivated action (4 items: e.g. ‘I am trying my best to accept my body’), and body criticism (4 items: e.g. ‘I am starting to think that I worry about my body too much’). The items pertaining to body criticism were re-coded to correct for negative scoring. Higher scores indicate greater body compassion (baseline α = 0.84, post-intervention α = 0.86).

Body and face satisfaction

The Face and Body cathexis scale (FBCS; Furnham & Greaves, 1994) asked participants to rate their satisfaction on 18 body and face parts and their overall appearance from 1 (extremely dissatisfied) to 5 (extremely satisfied). These included body parts such as ‘Nose’, ‘Bottom’, and ‘Tummy’. A high total score indicated greater face and body satisfaction (baseline α = 0.88, post-intervention α = 0.89).

Internalisation of thin-ideal

Participants completed the thin/low body fat subscale of the Sociocultural Attitudes Towards Appearance Questionnaire (SATAQ-4; Schaefer et al., 2015). This scale consists of five statements, rated on 5-point scales, 1 (definitely disagree) to 5 (definitely agree), for example, ‘I want my body to look very thin’. A higher score indicates greater internalisation of the thin-ideal (baseline α = 0.83, post-intervention α = 0.88).

Weight bias

Participants completed an amalgamated and modified version of the standardised weight bias scales informed by previous research (Stewart & Ogden, 2020). This 23-item scale was developed from the most commonly used standardised measurement scales of weight bias. It pools the items considered to be most reflective of conceptualisations of weight bias and uses person-first language. It has two subscales: negative stereotypes (15 items: e.g. ‘People with obesity lack willpower’), and dislike of people with obesity (8 items: e.g. ‘I really don’t like people with obesity much’). A higher score indicates higher weight bias (baseline α =0.93, post-intervention α = 0.95).

Behavioural intentions

Behavioural intentions were measured using a 9-item scale (Ogden & Arulgnanaseelan, 2017). Participants rated the extent to which they intend to engage in a series of behaviours on a scale ranging from 1 (not at all) to 5 (totally): intentions to eat healthily (3 items: e.g. ‘intend to eat more healthily’; baseline α = 0.77, post-intervention α = 0.81); intentions to exercise (3 items: e.g. ‘intend to be more active’; baseline α = 0.88, post-intervention α = 0.91); and intentions to manage their weight (3 items: e.g. ‘intend to lose weight’; baseline α = 0.84, post-intervention α = 0.87).

Procedure

This study is compliant with the University of Surrey’s procedure for conducting research with human subjects. Participants were able to complete this online study at home on any internet-enabled electronic device, via an anonymous weblink on the online survey platform host ‘Qualtrics’. Participants read an online information sheet, provided informed consent, and completed demographic and media attitudes measures before completing the health outcome measures. Participants were randomly allocated to one of the intervention conditions (body diversity, thin-ideal or control) and viewed the corresponding video. They then completed the health outcome measures again before being debriefed. Participants were free to terminate the study at any point by exiting the browser. The questionnaire took participants between 10 and 15 minutes to complete.

Data analysis

The data were analysed using SPSS Statistics (version 24). ANOVA and Chi-Square were run to describe participant demographics by condition and to check that the group randomisation had been successful. MANCOVA was run to assess for the impact of intervention group (body diversity, thin-ideal and control) on changes in outcome variables, whilst controlling for media literacy. Post-hoc tests were run on change scores for variables demonstrating statistically significant results to see where the differences lie, and to assess whether group differences were genuine. The de-identified data are openly available in Mendeley Data at doi.org/10.17632/vkyjyvdrjp.1.

Results

Participant demographics

Participant demographics for all participants and by intervention group are shown in Table 1.

Table 1. Participant demographics by condition

	All (N = 160)	Body diversity (N = 53)	Thin-ideal (N = 53)	Control (N = 54)	ANOVA values/ χ2 values
Age	M = 28.2 SD = 10.8 Range = 16–74	M = 27.8 SD = 10.1 Range = 16–59	M = 30.0 SD = 10.5 Range = 18–59	M = 29.7 SD = 11.7 Range = 17–74	F(2,157) = 0.63 p =.53
Ethnicity	White = 137 Black = 3 Asian = 9 Other = 11	White = 47 Black = 1 Asian = 4 Other = 1	White = 45 Black = 2 Asian = 2 Other = 4	White = 45 Black = 0 Asian = 3 Other = 6	χ^2(6) = 6.16 p =.41
BMI group	Underweight = 6 Healthy weight = 118 Overweight = 30 Obese = 6	Underweight = 2 Healthy weight = 40 Overweight = 8 Obese = 3	Underweight = 2 Healthy weight = 38 Overweight = 11 Obese = 2	Underweight = 2 Healthy weight = 40 Overweight = 11 Obese = 1	χ^2(6) = 1.65 p =.95

All participants were female (mean = 28.2 years). The majority were white and of a healthy weight. There were no differences by intervention group, suggesting the randomisation was successful.

The impact of intervention on outcome variables

The impact of the interventions of all outcome variables, whilst controlling for media literacy are shown in Table 2.

Table 2. Means, standard deviations, main effects and interaction effects of condition and time on outcome measures

	Body diversity (1) (N = 53)		Thin-ideal (2) (N = 53)		Control (3) (N = 54)
	Baseline M(SD)	Post-intervention M(SD)	Baseline M(SD)	Post-intervention M(SD)	Baseline M(SD)	Post-intervention M(SD)	Main effect of condition	Main effect of time	Condition x time
Body satisfaction	65.83 (12.68)	67.60 (12.70)	64.74 (11.12)	65.06 (12.18)	63.56 (12.63)	64.94 (13.09)	F = 0.65 p =.522 np^2 =.01	F = 2.49 p =.117 np^2 =.02	F = 1.30 p =.276 np^2 =.02
Body compassion	60.25 (9.74)	61.43 (9.96)	61.81 (9.56)	62.09 (11.14)	60.96 (8.78)	61.09 (8.94)	F = 0.20 p =.818 np^2 =.00	F = 0.98 p =.323 np^2 =.01	F = 0.89 p =.413 np^2 =.01
Internalisation of thin-ideal	16.13 (4.98)	15.32 (5.48)	15.32 (5.26)	15.34 (5.71)	14.37 (5.11)	14.22 (5.43)	F = 0.92 p =.403 np^2 =.01	F = 0.34 p =.563 np^2 =.00	F = 1.50 p =.227 np^2 =.02
Weight bias	51.25 (17.60)	45.62 (17.86)	48.00 (14.35)	46.26 (17.03)	53.26 (16.88)	50.20 (19.20)	F = 1.48 p =.235 np^2 =.02	F = 7.77 p =.006 np^2 =.05	F = 5.42 p =.005 np^2 =.07
Eating intentions	9.57 (2.64)	9.26 (3.06)	9.64 (2.74)	10.06 (2.76)	9.35 (2.69)	9.48 (2.75)	F = 0.44 p =.647 np^2 =.01	F = 1.01 p =.316 np^2 =.01	F = 3.69 p =.027 np^2 =.05
Exercise intentions	11.11 (2.59)	11.11 (2.83)	11.25 (2.73)	11.45 (2.69)	11.30 (2.58)	11.30 (2.56)	F = 0.12 p =.889 np^2 =.00	F = 2.79 p =.097 np^2 =.02	F = 0.35 p =.705 np^2 =.00
Weight management intentions	9.64 (3.40)	9.49 (3.42)	9.96 (3.54)	9.89 (3.63)	10.31 (3.25)	10.15 (3.13)	F = 0.53 p =.591 np^2 =.01	F = 0.22 p =.637 np^2 =.00	F = 0.08 p =.923 np^2 =.00

The results showed no significant effects of intervention group on changes in body satisfaction, body compassion, internalisation of the thin-ideal, intentions to exercise or intentions to manage weight (all p >.05). However, there were significant effects of intervention group on weight bias and intentions to eat healthily.

Weight bias post hoc tests

Change scores (post-intervention–baseline) were calculated and differences between groups were assessed using independent samples t-tests. Results showed that the decrease in weight bias was significantly greater for participants in the body diversity intervention group compared with the thin-ideal intervention group [t(104) = −3.45, p = .001, d = 0.67] and control group [t(105) = −2.23, p = .028, d = 0.43]. There was no significant difference in change scores between participants in the thin-ideal intervention group and control group [t(105) = 1.04, p = .300]. This indicates that those exposed to body diversity imagery reported a greater decrease in weight bias than those exposed to either the thin-ideal or control interventions.

Intentions to eat healthily post hoc tests

Change scores (post-intervention–baseline) were computed, and differences by group were assessed using independent tests. The results showed a significant difference in change scores for participants in the thin-ideal intervention group compared to the body diversity intervention group [t(104) = −2.55, p = .012, d = 0.50], but not between the body diversity intervention group and control group [t(98.96) = −1.59, p = .114] or between the thin-ideal intervention group and control group [t(105) = 1.15, p = .253]. This indicates that those exposed to thin-idealistic imagery reported a greater increase in intentions to eat healthily than those exposed to body diversity intervention.

Discussion

Although a growing body of literature suggests the efficacy of changing the type of media imagery (Diedrichs & Lee, 2011; Halliwell et al., 2005; Hendrickse et al., 2020; Ogden et al., 2020), the impact of this on weight bias has not yet been fully explored. This study therefore aimed to evaluate the comparative impact of both body diversity and thin-ideal interventions on health outcomes, including body satisfaction, body compassion, internalisation of the thin-ideal, weight bias and behavioural intentions to eat more healthily, exercise more and successfully manage weight, compared to a control group, whilst controlling for media literacy.

Findings demonstrated that those exposed to body diversity imagery showed significantly lower levels of weight bias post-intervention compared with baseline and those in the other intervention groups. This finding supports prior research demonstrating that repeated exposure to larger body sizes can shift perceptions on body size preferences, and the tolerance and acceptance of obesity (Boothroyd et al., 2012; Robinson & Kirkham, 2014). This therefore supports calls for the media presenting more diverse and realistic images, and indicates that a more diverse presentation of attractiveness ideals could be helpful to reduce stigmatising attitudes towards those with surplus body weight. Through processes of internalisation and upward social comparisons, sociocultural theory contends that thin-idealistic media messaging creates and exacerbates negative body appraisals (Thompson et al., 1999; Tiggemann & McGill, 2004; Tiggemann & Miller, 2010). Sociocultural theory can therefore be useful in understanding how reducing thin-ideal discourse may be effective in reducing weight bias. Pressure to conform to thin-ideals can lead to the internalisation of anti-fat and stigmatising attitudes towards those with overweight and obesity through the implicit messaging that thinness equates to happiness, success and gain (Rodgers, 2016). Therefore, removing the thin-idealistic messaging and replacing it with body diversity images could start to challenge stigmatised attitudes towards body weight.

Weight normalisation theories can also be helpful to understand these findings. An increasing body of research has demonstrated that exposure to a broader range of body sizes leads to the normalisation of larger body weights (Oldham & Robinson, 2016; Robinson & Kirkham, 2014). Internalisation could therefore function in a more constructive way. Increased exposure to a range of diverse imagery that challenges stereotypical thin-ideal beauty standards, presented in an affirmative way, could lead to the internalisation of a more diverse sense of what is beautiful. This could therefore lead to individuals appraising both the bodies of themselves and of other people too, in a more positive light.

The results also showed that those who participated in the ‘thin-ideal’ intervention group had significantly greater intentions to eat healthily compared with baseline and those in the other intervention groups. It has previously been suggested that body dissatisfaction might not always be universally deleterious; some degree of negative body image could be helpful to motivate individuals towards engaging in health behaviours (Heinberg et al., 2001). Whist this view has been widely refuted (Puhl et al., 2007; Tylka & Wood-Barcalow, 2015), the results of the present study suggest that some degree of exposure to thin-idealistic media messaging increases intentions to eat healthily, whereas exposure to body diversity reduces these intentions. It would therefore be important for future research to be conducted to determine at what point thin-idealistic imagery becomes damaging to consumers’ motivations to engage in health behaviours, by addressing the complex relationship between body dissatisfaction, and motivation and participation in health behaviours. Furthermore, whether this translates to actual eating behaviour or not needs to be determined before firm conclusions on this are drawn.

There were no significant effects of the interventions on exercise intentions or changes in body satisfaction contrary to previous research (Hendrickse et al., 2020; Mulgrew et al., 2018). There were also no significant effects for body compassion, internalisation of the thin-ideal and weight management intentions, contrary to predictions. It is possible that for these variables, a 1-minute intervention video was not sufficient time to elicit an effect.

Findings from this study hold important implications for improving the wellbeing and body confidence of all women who are exposed to media. In particular, these findings lend empirical validation that campaigns such as ‘This Girl Can’ are successful in promoting body diversity and reducing the stigmatisation of those with larger body sizes. Increasing body diversity across the media would be beneficial for the reduction of weight bias, which in turn could improve general wellbeing and preclude the development of body image problems (Bury et al., 2016; Derenne & Beresin, 2018; Halliwell et al., 2005; Mayer-Brown et al., 2016; Moreno-Domínguez et al., 2019; Pedersen et al., 2018; Robinson et al., 2017; Slevec & Tiggemann, 2011). Body diversity could be integrated into existing programmes to manage problems with body image and eating behaviour.

It is important to note that this study is not without limitations. Firstly, the majority of the female participants in this study were white, which could restrict the generalisability of these findings. This is relevant given that many of the women in both the body diversity intervention and thin-ideal intervention represent those from black and minority ethnic backgrounds. Second, the age range of the sample was wide and included participants up to 74 years. Given that the videos used are available to all ages the aim was to test their impact on their ‘natural’ audience. However, it is acknowledged that this may have reduced the capacity for social comparisons given that the age of the women portrayed in the videos was younger than this. Third, it is possible that the participants could have guessed the aims of the study, and therefore presented more socially desirable responses. Fourth, whilst the interventions emphasised being active the aspects of body image assessed tended to focus on static components. Finally, this study only evaluated the impact of short-term exposure to each of the intervention groups. Further research could therefore include a more diverse sample, directly assess the role of social comparisons, broaden the notion of body image and assess the repeated impact of such interventions over a longer period of time.

In conclusion, although previous research has highlighted the potential impact of changing the type of imagery used in the media on measures of body appraisal and behaviour, research has not yet investigated this with regards to weight bias. Findings indicated a degree of clinical equipoise; whereby exposure to body diversity images reduced weight bias, and exposure to the thin-ideal promoted intentions towards healthy eating. These findings therefore provide empirical evidence for using different types of media imagery in order to change different health outcomes.

Disclosure statement

The authors declare that there is no conflict of interest.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.