Diversity and inclusion in images

How to choose images

Background:

Images convey an organization’s values and its intended or desired audience. Because the chemistry community is and should be diverse, images that reflect that diversity are more accurate and will connect more with a chemistry audience. When people don’t see themselves in images, they feel invisible and may even perform worse on tasks (J. Soc. Psychol. 2010, DOI: 10.1080/00224540903366552). 

While it is important to consider various aspects of diversity in images, trying to represent all races, gender identities, ethnicities, etc. in every image is impossible and could seem unnatural. A more thoughtful approach will depict more authentic and realistic images.

Recommendation:

When choosing images, look for ones that reflect the diversity of the audience you have or want to have. The images should also match your content. Characteristics to think about varying in images include race and ethnicity, skin tone, gender identity and expression, age, disability status, body size and shape, and hair texture. When searching for stock images, use specific terms and descriptions so that you can find appropriate images. For example, instead of just “scientist,” “lab worker,” and “teacher,” try “scientist in wheelchair,” “Hispanic lab worker,” and “Black female teacher smiling.”

Do not attempt to show every aspect of diversity in every image (see “Performative diversity in images”). Instead, be thoughtful about what makes sense for your content, its purpose, and your audience. For example, in an image of multiple people, consider whether you can include at least some varied characteristics (e.g., race, gender, age), and make sure that when you choose an image of a single person, you’re not always depicting the same attributes in the same combinations. Seek guidance and feedback from others to evaluate whether your images are depicting diversity and are inclusive. In addition, involve a diverse group of people in the process of choosing, creating, and editing images for your content (see “Involve a diverse group of people in the creative process”).

Examples of websites that focus on increasing the representation of marginalized groups in photos and illustrations include the following:

General

  • Humaaans: A mix-and-match tool for creating images with different skin tones, hair, clothes, and disabilities (free)
  • Thenounproject.com: Icon and stock photos that celebrate diversity (free)

Age

Body size

Disability

Gender

Race and ethnicity

Example:

Use:

A collage of various scenes from a Black student's experience, including publishing, researching, and teaching.
Ryan Inzana

Illustration for an article on graduate school in the sciences (C&EN, Sept. 9, 2018)  

Performative diversity in images

Background:

Superficial displays of diversity and inclusion in images imply a lack of commitment to these ideals and can indicate a desire to deceive the audience or to pursue diversity for only personal or business gain rather than fair representation and true change. For example, having an image of people of color on the cover of a textbook but not inside is a form of bias (J. Chem. Educ. 2007, DOI: 10.1021/ed084p342). And choosing an image showing diversity that does not match the topic or trying to show every type of race in a single image may seem as if diversity and inclusion efforts are just a box-checking activity and not a thoughtful, intentional process. For example, if an article is about Black scientists, an image that shows many different races might not be appropriate. Instead, an image that shows the diversity of Black people would be a better choice.

Recommendation:

Before choosing an image for content, think carefully about what the image should accomplish, the topic covered, and the intended audience. Fully commit to making diversity and inclusion considerations part of the art process rather than thinking of diversity as a stand-alone factor that is taken into consideration only toward the end of content creation. For example, consider reorganizing the art process to think about diversity and inclusion at the beginning of every project, and recruit a diverse team of people to help create and choose images.  

Examples:

Use:

In a video aiming to recruit new employees, having the president or CEO acknowledge the current lack of diversity and explain what they are doing to make the environment more welcoming to all employees

Avoid:

Hiring a diverse cast to act as employees because the company doesn’t have a diverse workforce

Use:

Consistently showing Black people in images on all platforms throughout the year

Avoid:

Including images of Black people on social media only during Black History Month

Stereotypes in images

Background:

Just because an image shows diversity or a person from an underrepresented group does not mean that it is respectful or inclusive. Some images perpetuate stereotypes through the subjects’ actions, dress, environment, or positioning. For example, in a science textbook, depicting Black people only as athletes and not as scientists perpetuates stereotypes (J. Chem. Educ. 2007, DOI: 10.1021/ed084p342). Other examples of stereotypes in images include showing photos of transgender people before and after they transition and showing stereotypically gendered imagery of trans people. Publishing these kinds of images reinforces biases and disrespects people. Learning about your audience is one way to recognize and avoid harmful stereotypes in images.

Recommendation:

Examine how people are shown in images, and evaluate whether the images reinforce harmful stereotypes. Some aspects to think about when reviewing images include people’s positioning, such as who is above or supported by others; people’s actions, including whether people are portrayed as actors or bystanders; and people’s clothing. Do not choose images that perpetuate stereotypes or that have potentially demeaning portrayals of specific groups. 

Have a range of people review images for signs of bias, especially if the image is related to a cultural group outside your own (see “Involve a diverse group of people in the creative process”). When searching for images that are inclusive and respectful, look for databases by organizations committed to respectful portrayals of people. For example, several organizations have image banks that avoid stigma around body size. 

Example:

Use:

In an article on body size, an image that shows a full person (not just their stomach) doing an everyday activity

Avoid:

An image that shows just a stomach or a person eating messily

Accuracy in images

Background:

Images send powerful messages. Images that contain inaccuracies can do harm, especially when the images are of people or related to people marginalized because of an aspect of their identity. Beyond the content in an image, an image’s placement in relation to text may create an error if it incorrectly connects text to an image. 

Recommendation:

Ensure images are accurate in terms of their content and placement in relation to other images and text. For example, if depicting a specific culture, consider whether the image is culturally accurate.

Example:

Use:

Pairing an image of an individual with a quote by that person

Avoid:

Pairing a quote from someone with a stock image of a different person

Editing photos

Background:

How photographs are edited can greatly affect how the audience perceives them. In addition, editors that have to crop photos may decide who gets to be seen by a wide audience. Photo illustrations—images that include altered photos or photos combined with illustrations—are also prone to bias and can perpetuate stereotypes, especially if a photo’s manipulation is not readily perceivable or disclosed to the audience. Editing photos to artificially show more diversity (such as by using Photoshop to include more people of color in a photo) can also be inauthentic and treat diversity like a marketing tactic instead of a true value and is not condoned by the American Chemical Society. 

An additional consideration is using color and brightness correction for people with different skin tones. Sarah Lewis explains in a 2019 New York Times article, “Light skin became the chemical baseline for film technology.” Properly adjusting images for different skin tones conveys respect for people and ensures fair and accurate representation in images.

Recommendation:

Follow your organization’s ethical principles on editing photographs. ACS doesn’t manipulate images in a way that will mislead or misrepresent people. Carefully consider how to crop images—who is being made visible, and who remains invisible? And when making photo illustrations, ensure that the way photos are edited do not perpetuate stereotypes (see “Stereotypes in images”) or commodify diversity by exploiting marginalized identities for gain. In addition, editors should take special care to accurately depict darker skin tones in photographs.

Skin tones in illustrations

Background:

All In, a toolkit by Google, says, “Across the board, people find illustrations with orange or yellow skin tones to represent white users.” For example, Lego has gotten pushback for saying yellow is a skin color that could represent anyone. “LEGO figurines reflect the norms of a dominant culture under the guise of neutrality,” Samantha Allen writes in the Daily Beast.

Recommendation:

When possible, choose human skin tones for human figures in illustrations. If it is important to use nonhuman skin tones, show a variety of shades of color to simulate the variety of skin tones in real life.

Captions for images

Background:

Captions tell the audience who and what is happening in an image. Failing to mention someone or incorrectly identifying someone may be an honest mistake, but it may also be the result of unconscious bias and can contribute to people’s feelings of exclusion. Consider the common experience among Asians of being mistaken for someone else of the same race. Misidentifying people in a photo can imply that you don’t truly see or care about them. 

In addition, using different words to describe the same action by different people can show bias (e.g., using “looting” to describe a Black person but “finding” for White people in photos after Katrina).

Recommendation:

Carefully consider who you name in a caption and how you do it. Generally, everyone should get equal treatment—for example, full names for everyone. If you do not know everyone’s name or their name is not important to the audience’s understanding, still acknowledge that other people are in the image (e.g., “team members,” “lab mates”). If you give the job title or another piece of information for one person but not another, there should be a good reason.

Example:

Use:

Sir Martyn Poliakoff seated with eight young people behind him.
Several 2019 CAS Future Leaders pose with Sir Martyn Poliakoff (front) of the University of Nottingham, who is known for his periodic table videos.
Linda Wang/C&EN

An image caption acknowledges the role of everyone in the photo (C&EN, Sept. 14, 2019).

Resources on inclusive images

Previous page

Job descriptions

Next page

Data visualization

The ACS takes your privacy seriously as it relates to cookies. We use cookies to remember users, better understand ways to serve them, improve our value proposition, and optimize their experience. Learn more about managing your cookies at Cookies Policy.