Enhancing comprehension of online informed consent: the impact of interactive elements and presentation formats

ABSTRACT

Informed consent, a cornerstone of research ethics, ensures participant protection and informed participation, particularly in online settings. Despite its significance, engagement with online consent forms remains low, underscoring the need for improved presentation strategies. This study investigates the impact of interactive elements and diverse presentation formats on the comprehension and engagement of online informed consent documents among a broad demographic beyond the commonly studied student populations. Employing a between-subjects experimental design, we explored six versions of online consent forms varying in interactivity, readability, and visual formatting to identify optimal strategies for enhancing participant comprehension and engagement. Our findings reveal that interactive formats significantly improve comprehension and perceived readability, highlighting the pivotal role of design in facilitating informed consent. The study also examines the influence of individual differences, such as self-efficacy and trust in science, on the effectiveness of consent forms, providing insights into the nuanced dynamics between participant characteristics and consent form engagement. These results advocate for integrating interactive elements and thoughtful design in consent forms to foster a more informed and engaged participant base. Implications for research ethics, best practices in consent form development, and future research directions are also discussed, emphasizing the need for ongoing innovation in the consent process to adapt to the evolving landscape of online research. This study contributes to the body of knowledge on research ethics by offering evidence-based recommendations for enhancing the informed consent process, ultimately promoting participant-centered research practices.

KEYWORDS:

• Informed consent
• research ethics
• comprehension
• best practices

INTRODUCTION

Informed consent is a critical exchange between researchers and potential participants in the research process. This process aims to protect the research participants by providing information to the potential participants about the possible risks, benefits, and details of the research study to help them make an informed decision about their participation (Nijhawan et al., 2013). Despite the value of this process, previous studies demonstrate that most participants do not thoroughly read consent forms (Douglas et al., 2021; Perrault & Keating, 2018; Rao, 2008; Research, N. C. for the P. of H. S. of B. and B, n.d.; Varnhagen et al., 2005). Although previous research has sought to understand why participants do not thoroughly read consent forms, many of these studies have limited their inquiry to population samples that only include university students (Geier et al., 2021; Perrault & Keating, 2018; Theiss et al., 2014). However, student samples do not reflect the general population as they tend to be more homogenous than representative samples (Hanel et al., 2016). Thus, to adequately address the potential risks that participants may experience during the research process and improve consent forms, we need to examine this phenomenon among a broader population.

LITERATURE REVIEW

Overall, studies on informed consent forms show that university students commonly refrain from thoroughly reading or comprehending consent forms, especially in online research (Geier et al., 2021; Perrault & Keating, 2018; Perrault & McCullock, 2019). In Perrault and Keating’s study (Perrault & Keating, 2018) participants reported that if consent forms were shorter and presented only essential information, the readability would be improved, suggesting that they would be more likely to read them. Previous research documents the negative impacts of having lengthy consent forms (Corneli et al., 2017; Emanuel & Boyle, 2021). This research suggests that consent forms that are too long impede participants’ understanding of these forms. Due to the prevalence of long consent forms, federal guidelines via the “Common Rule” were implemented in 2017 that require consent forms to be concise and to promote comprehension (Office for Human Research Protections, 2021). Applying the Common Rule to the informed consent process helps facilitate comprehension. The Common Rule defines comprehension as providing information that a “reasonable person” would understand, enabling them to decide to participate in the research study. However, despite this implementation, Perrault and McCullock (2019) show that while participants reported enjoying a more streamlined consent form process as suggested by the Common Rule, on average, participants understood only 50% of the information included in these shortened forms. Although these previous studies and new federal guidelines help better develop consent forms by improving how positively participants respond to the forms, their results suggest that participants are still not thoroughly reading the forms, which questions whether researchers adequately consent participants into the study. If the information presented in the consent form is not comprehensible, as defined by the Common Rule, the ethical commitment to protect participants has not been upheld (Research, N. C. for the P. of H. S. of B. and B, n.d.).

More recent research has implemented techniques recommended by past studies to enhance the readability and comprehension of online consent forms (Douglas et al., 2021; Geier et al., 2021; Karegar et al., 2018; Lindegren et al., 2019; Perrault & Keating, 2018; Perrault & McCullock, 2019). Readability, which refers to the ease with which a reader can understand written text, plays a crucial role in ensuring that the information is conveyed clearly and in a way that a varied audience can understand. Key factors influencing readability include the text’s difficulty level, as more complex sentences may take the average reader longer to process. Comprehension is equally vital in the context of consent forms. While it shares similarities with readability, comprehensibility is influenced not only by the difficulty of the text but also by various reader characteristics. These characteristics encompass the reader’s background, prior knowledge and understanding of the topic, and interest in the subject matter. Additionally, an author can enhance comprehension by organizing ideas logically and coherently, allowing the content to flow smoothly. Furthermore, the findings from Geier et al.‘s study demonstrated that highly interactive consent forms, such as those structured with brief questions, significantly improved comprehension and readability scores among participants compared to other form variations (Geier et al., 2021). These results underscore the importance of modifying the presentation of the consent form and encouraging participants to interact with it to foster a better understanding of consent forms.

While these previous studies provide valuable insights into improving participants’ comprehension of the online informed consent process among student populations, their recommendations may not extend to participants outside academic settings (Geier et al., 2021; Perrault & Keating, 2018; Perrault & McCullock, 2019). Therefore, to address this gap, the current study explores the comprehension of informed consent forms within a more representative U.S. population. Additionally, the study seeks to understand how individuals’ perceptions of self-efficacy and trust in science influence this process. Researchers can effectively convey their intended message and key points by finding the right balance between readability and comprehension while ensuring that participants fully grasp and retain the information presented in consent forms.

RESEARCH QUESTIONS

The focus of this study is to assess participants’ comprehension and perception of readability (how easy a consent form is to read) of multiple versions of the consent form. Previous studies on online consent forms have tested variations in the presentation of consent forms, such as bold and highlighted text, bullet points, and other means of visually formatting text and the reading level (Perrault & Keating, 2018; Perrault & McCullock, 2019). Other studies have looked at interactive elements to draw participants’ attention to important content, with mixed results (Lindegren et al., 2019; Lustria, 2007; Rowbotham et al., 2013). This study assesses six versions of the same consent form (Table 1), which include variations on the visual formatting of text, interactivity, and reading level, to answer the following research questions:

Table 1. Consent form variations.

Form	Description	Words	Participants (%)
Control	Standard consent form, language adapted from IRB boilerplate	466	111 (14.8)
Special Formatting	Standard consent form with the use of bullet points, bold text, and highlighted words	466	114 (15.2)
Short	Word count significantly reduced, with option to view full consent form	89	119 (15.9)
4th Grade Reading Level	Consent form language revised to a 4th grade reading level	375	120 (16.0)
Interactive Control	Standard consent form, split over 5 pages with questions about the study at the end of each page	466	130 (17.4)
Interactive 4th Grade Reading Level	4th grade reading level version, split over 5 pages with questions about the study at the end of each page	375	155 (20.7)

(RQ) 1:

(a)Which version of the consent form has the highest level of comprehension by the participants? (b):Does interactivity improve participants’ comprehension of consent forms?

RQ 2:

Which consent version is perceived as the easiest to read?

Another goal of this study is to assess whether specific individual differences affect whether participants read consent forms. Bandura (1989) defines self-efficacy as an individual’s belief about their capability to influence situations that impact their lives. Thus, an individual’s levels of perceived self-efficacy could have implications for their assessment of the adequacy of the information provided, their perception of risk, and the actions the individual believes they can take during the study because of reading an informed consent document (Donovan Kicken et al., 2012; Geier et al., 2021). A previous study found that self-efficacy was not significant in whether college student participants read consent forms (Geier et al., 2021). However, student populations will likely have higher self-efficacy regarding research participation than the general population. These interactions cause students to have a different relationship with academic research than persons outside the academy (Hanel et al., 2016. Thus, it is essential to consider this variable with a general population. Therefore, we ask the following research question:

RQ 3:

Do the participants’ perceptions of self-efficacy relate to a) the level of comprehension of the consent form and b) the perceived readability of the consent form?

Additionally, this study aims to investigate the crucial role of an individual’s level of trust in science and researchers in shaping their engagement with consent forms. The historical disenfranchisement and harm experienced by marginalized groups at the hands of the government and researchers have underscored the need for rebuilding trust and ensuring ethical research practices (Ogungbure, 2012; Washington, 2006; Whitehead et al., 2016). Furthermore, skepticism toward science and research methods has been documented as a growing concern in the research literature (Benegal & Scruggs, 2018; Holman & Lay, 2018; Wynne, 2006). Notably, a recent report by the Pew Research Center suggests that Americans, as a whole, exhibit low levels of trust in scientists (Kennedy et al., 2022). While a previous study found no significant connection between participants’ trust in researchers and their perceptions of consent forms, it is important to note that this study was limited to university students (Geier et al., 2021). Consequently, extending this assessment to populations outside the academic setting is crucial to gain a more comprehensive understanding of the influence of trust in researchers on the informed consent process. By investigating the influence of participants’ perceptions of researchers on their comprehension and perceived readability of consent forms, this study contributes to a more comprehensive understanding of the factors that shape individuals’ engagement with the informed consent process. These insights can inform the development of strategies to enhance the clarity, accessibility, and effectiveness of consent forms for diverse populations. Therefore, we pose the following research question:

RQ 4:

Do the participants’ perceptions of researchers impact a) the level of comprehension of the consent form and b) the perceived readability of the consent form?

MATERIAL AND METHODS

Research design

Our study employed a between-subjects experimental design, chosen for its ability to effectively address our research questions regarding the impact of consent form presentation on comprehension. We designed the study to examine the impact of different consent form styles, including interactivity and reading level, on participants’ comprehension and perception of the forms. Key elements of our design included the development of six variations of the same consent form to test, including interactive formats that engaged participants through embedded questions.

Participants

A total of 749 participants were included in the analysis. We performed a preliminary power analysis based on previous studies (Perrault & Keating, 2018; Perrault & McCullock, 2019). With an alpha = .05 and power = .95, the projected sample size needed with this effect size (GPower 3.1) is approximately N = 714. Participants were recruited through Dynata, a population recruitment service. The population sample used for this study is a non-probability sample. Participants were recruited based on their representativeness of the U.S. adult population for identity and demographic characteristics, including gender, race/ethnicity, and education level. Data from participants who did not complete the survey and did not correctly answer the attention check question were excluded from the results. In return for completing the survey, participants earned payment in U.S. dollars. By utilizing the participant recruitment service, participants remained anonymous to researchers. This study was declared exempt by the university’s institutional review board (IRB).

Procedures

Once participants opted into the online survey, they were randomly assigned to one of six versions of the same consent form: 1) Control, 2) Interactive 4^th Grade Reading, 3) 4^th Grade Reading Level, 4) Short, 5) Special Formatting, and 6) Interactive Control (see Table 1). All versions of the consent forms included content explaining the study’s purpose, risks, benefits, voluntary participation, details of the survey’s structure, information about confidentiality and privacy of the participants’ data, compensation for participating in the study, and contact information for the primary investigator (PI) and the IRB at the research team’s university. Participants were informed that the purpose of the study was to evaluate different formats of informed consent forms. Informing participants about the study’s focus on informed consent may prompt them to read the forms more closely than they would otherwise. The six versions of the consent forms were included to extend findings from previous research surrounding formatting and language usage in consent forms and the effects of interactivity on comprehension (Douglas et al., 2021; Geier et al., 2021; Karegar et al., 2018; Lindegren et al., 2019; Perrault & Keating, 2018; Perrault & McCullock, 2019).

Measures

After the participants reviewed the consent form version that they were randomized to, they answered comprehension questions about the content of the consent form, were asked to what extent they had read it, and provided their perceptions of the form’s readability (Perrault & Keating, 2018). Comprehension was measured using seven questions such as, “Who is the primary investigator of this study?” Participant responses were categorized as either incorrect = 0 or correct = 1. Therefore, a score of 7 is equivalent to a perfect score on comprehension. Additionally, readability was assessed by asking participants to report how easy or difficult they found the consent form to read on a 10-point scale where 1 = Very Easy and 10 = Very Hard.

Participants were also asked questions about self-efficacy and their level of trust in researchers and scientists. Self-efficacy (ɑ = .90) was assessed using a 10-item validated scale, which included items like, “I can manage to solve difficult problems if I try hard enough,” “I am confident that I could deal efficiently with unexpected events,” and “When I have a problem, I can find several ways to solve it” (Hong et al., 2016). Participant responses were based on a 5-point scale where 1 = Never and 5 = Very Often. Trust in researchers and scientists was evaluated using an 18-item validated scale (ɑ = .87) (Critchley, 2008), including items such as “[Scientists/researchers] Enjoy being treated as important people,” “[Scientists/researchers] Want to make life better for ordinary people,” and “[Scientists/researchers] Are highly trained in what they do.” The responses were based on a Likert scale where 1 = Strongly Agree and 5 = Strongly Disagree. In addition to assessing comprehension and readability, we also measured the time participants took to complete the consent forms.

Two open-ended questions asked participants to self-report why they had not read the form (if they indicated they had not) and gather feedback on what they felt would encourage more people to read consent forms. Last, participants answered basic demographic questions, such as age, gender, level of education, and race/ethnicity.

Analysis

Statistical analyses were conducted using SPSS (v.26). Descriptive statistics were performed to summarize the sample characteristics and study variables. An analysis of variance (ANOVA) test was conducted to evaluate statistically significant between-group differences between consent form conditions. Post-hoc Tukey analysis was performed to identify the highest comprehension condition. In addition, linear regression analyses were run to explore significant predictors of comprehension and readability.

Qualitative analysis was conducted on the participant responses to two open-ended questions using a descriptive analysis approach (Braun & Clarke, 2006). Initial thematic codes were developed based on participant responses to the same questions from a previous study (Geier et al., 2021). Two research team members iteratively revised the codes as needed. Both members then tested the final codes using 10% of the responses, with an inter-rater reliability of 94.5% for the question regarding the reasons for not reading the consent and 83.6% for suggestions on making consent forms better. Once inter-rater reliability was established, the remainder of the responses were coded individually, and the resulting codes were then added to SPSS for further descriptive analysis. Both research team members highlighted responses that contained more nuance than the codes could account for during the coding process. Further qualitative analysis was performed on these responses to examine directions for future research.

RESULTS

Participants were nearly evenly randomized to each of the groups: Control (n = 111, 14.8%), Special Formatting (n = 114, 15.2%), Interactive 4^th Grade Reading level (n = 155, 20.7%), 4th Grade Reading Level (n = 120, 16%), Interactive Control (n = 130, 17.4%), and the Short version (n = 119, 15.9%). A majority of the participants were older adults over the age of 55 (52.7%). Most participants were males (53.9%) and White (75.2%). See Table 2 for demographic details.

Table 2. Participant demographics.

Variable	Frequency (%)
Group	Special Formatting	Short	4^th Grade Reading Level	Interactive Control	Interactive 4^th Grade Reading Level	Control	Total
N	114	119	120	130	155	111	749
Age (years)
18–24	12 (10.5)	8 (6.7)	9 (7.5)	5 (3.8)	5 (3.2)	7 (6.3)	46 (6.1)
25–34	11 (9.6)	7 (5.9)	12 (10)	15 (11.5)	15 (9.7)	9 (8.1)	69 (9.2)
35–44	17 (14.9)	15 (12.6)	18 (15)	16 (12.3)	24 (15.5)	13 (11.7)	103 (13.8)
45–54	16 (14)	26 (21.8)	15 (12.5)	32 (24.6)	26 (16.8)	22 (19.8)	137 (18.3)
55–64	28 (24.6)	35 (29.4)	26 (21.7)	25 (19.2)	32 (20.6)	29 (26.1)	175 (23.4)
65–74	26 (22.8)	25 (21)	31 (25.8)	34 (26.2)	43 (27.7)	25 (22.5)	184 (24.6)
75+	4 (3.5)	3 (3.5)	9 (7.5)	3 (2.3)	10 (6.5)	6 (5.4)	35 (4.7)
Level of Education
No schooling completed	1 (.9)	0 (0)	2 (1.7)	0 (0)	0 (0)	0 (0)	3 (.4)
Kindergarten to 8th grade	0 (0)	0 (0)	0 (0)	1 (.8)	1 (.6)	1 (.9)	3 (.4)
Some high school, no diploma	0 (0)	2 (1.7)	3 (2.5)	1 (.8)	0 (0)	4 (3.6)	10 (1.3)
High school graduate with diploma or GED	8 (7)	17 (14.3)	12 (10)	11 (8.5)	16 (10.3)	13 (1.7)	77 (10.3)
Some college, no degree	21 (18.4)	16 (13.4)	13 (10.8)	14 (10.8)	23 (14.8)	12 (10.8)	99 (13.2)
Trade/technical/vocational certification	3 (2.6)	6 (5)	3 (2.5)	7 (5.4)	9 (5.8)	8 (7.2)	36 (4.8)
Associate degree	9 (7.9)	15 (12.6)	12 (10)	9 (6.9)	10 (6.5)	8 (7.2)	63 (8.4)
Bachelor’s degree	33 (28.9)	33 (27.7)	45 (37.5)	52 (40)	55 (35.5)	33 (29.7)	251 (33.5)
Master’s degree	21 (18.4)	18 (15.1)	23 (19.2)	24 (18.5)	32 (20.6)	20 (18)	138 (18.4)
Professional degree	7 (6.1)	6 (5)	5 (4.2)	7 (5.4)	6 (3.9)	3 (2.7)	34 (4.5)
Doctorate degree	9 (7.9)	6 (5)	2 (1.7)	4 (3.1)	3 (1.9)	9 (8.1)	33 (4.4)
Other	2 (1.8)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	2 (.3)
Gender Identity
Man	68 (59.6)	61 (51.3)	67 (55.8)	65 (50)	81 (52.3)	62 (55.9)	404 (53.9)
Woman	45 (39.5)	58 (48.7)	53 (44.2)	63 (48.5)	74 (47.7)	48 (43.2)	341 (45.5)
Trans woman	1 (.9)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1 (.1)
Genderqueer	0 (0)	0 (0)	0 (0)	1 (.8)	0 (0)	0 (0)	1 (.1)
Other	0 (0)	0 (0)	0 (0)	1 (.8)	0 (0)	1 (.9)	2 (.3)
Ethnicity origin (or Race)
White	93 (81.6)	103 (86.6)	99 (82.5)	111 (85.4)	129 (83.2)	90 (81.1)	625 (74)
Black or African American	11 (9.6)	7 (5.9)	10 (8.3)	9 (6.9)	13 (8.4)	10 (9)	60 (7.1)
American Indian or Alaska Native	3 (2.6)	4 (3.4)	4 (3.3)	0 (0)	1 (.6)	1 (.9)	13 (1.5)
Asian	9 (7.9)	8 (6.7)	11 (9.2)	7 (5.4)	7 (4.5)	8 (7.2)	45 (5.3)
Native Hawaiian or Pacific Islander	0 (0)	1 (.8)	2 (1.7)	0 (0)	1 (.6)	0 (0)	4 (.5)
Spanish, Hispanic or Latino/a	15 (13.2)	16 (13.4)	9 (7.5)	12 (9.2)	13 (8.4)	6 (5.4)	71 (8.4)
Other	5 (4.4)	6 (5)	3 (2.5)	4 (3.1)	6 (3.9)	3 (2.7)	27 (3.2)

Note: Participants were asked to check all that apply for ethnicity origin or race.

Comprehension and readability were correlated (r = −.25, p < .001). Our analysis revealed no significant differences in the time to complete the survey across the different consent form conditions. This finding indicates that participants allocated a similar amount of time, on average, to complete the survey regardless of the specific content of the consent forms. There were also no significant differences between the randomized groups regarding time (minutes) spent taking the survey, F(5, 352) = 1.26, p = .28; Control (M = 7.72, SD = 5.42), Interactive 4^th Grade Reading Level (M = 9.11, SD = 5.89), 4^th Grade Reading Level (M = 8.33, SD = 4.51), Short (M = 8.22, SD = 4.78), Special Formatting (M = 8.52, SD = 11.17), and Interactive Control (M = 10.69, SD = 9.71). However, it is important to note that the time taken to complete the survey may not solely reflect the engagement with and understanding of the consent forms. Factors such as reading speed, individual differences in attention or motivation, and external distractions could have influenced the variability in completion time (Malhotra, 2008).

Research question 1a-b

The first research question asked which version of the consent form had the highest level of comprehension among participants. ANOVA results for comprehension suggest a significant difference in comprehension between the various versions of the consent form F(5, 743) = 95.233, p < .001. In exploring if interactivity improved comprehension (RQ1b), post-hoc Tukey tests revealed those participants in the page-by-page conditions (Interactive Control and Interactive 4^th Grade Reading Level had significantly higher levels of comprehension compared to each of the other groups (see Table 3). There were no other significant differences between conditions. See Table 3 for the means and standard deviations for all variables by condition. See Table 4 for the number of participants in each group who had a perfect score on comprehension.

Table 3. Means and standard deviations for variables by condition.

		Comprehension	Readability
	N	M (SD)	M (SD)
Control	111	2.17a (1.40)	6.04 (2.85)
Special Formatting	114	2.22a (1.53)	5.96 (3.00)
Short “Brevity”	119	2.50a (1.60)	5.97 (2.83)
4th Grade Reading Level	120	2.26a (1.70)	6.38 (2.74)
Interactive Control	130	4.57b (1.53)	5.21 (2.78)
Interactive 4th Grade Reading Level	155	4.90b (1.30)	5.07 (2.77)
Total	749	3.22 (1.93)	5.73 (2.86)

Comprehension was scored on a scale from 0 to 7, where 7 was the highest score a participant could receive. Readability was ranked from 1 to 10 where 1 was the most readable.

Note: Means with different subscripts in the Comprehension column differ at p < .01 using the post-hoc Tukey test.

Table 4. Perfect scores for comprehension.

Group	n
Control	0
Special Formatting	2
Short “Brevity”	1
4^th Grade Reading Level	1
Interactive Control	2
Interactive 4th Grade	6

Research question 2

This research question asked whether any one version of the consent form was perceived as easier to read than the others. ANOVA results suggest a significant difference in readability between the various versions of the consent form, F(5, 743) = 4.462, p = .001. Post-hoc Tukey tests revealed that participants in the Interactive 4^th Grade Reading Level and the Interactive Control had significantly higher readability levels than the 4th-Grade Reading Level (see Table 5). There were no other significant differences between conditions.

Table 5. Significant post-hoc tukey test results for readability.

Group	Group	Mean Difference	Std. Error	p
4th Grade Reading Level	Special Formatting	.418	.369	.868
	Interactive 4th Grade	1.312*	.343	.002
	Interactive Control	1.176*	.357	.013
	Control	.347	.372	.938
	Short “Brevity”	.409	.365	.874

Research question 3

Results from linear regression suggest that levels of self-efficacy did have an effect on the level of comprehension of the consent form, F(1, 747) = 15.63, p < .001, R² = .02, and perceived readability, F(1, 747) = 14.784, p < .001, R² = .02. Meaning that increased self-efficacy resulted in higher comprehension and perceptions of readability of the informed consent. See Table 6.

Table 6. Regression Model for effect of self-efficacy on comprehension and readability.

	b	SE	β	95% CI		p
				LL	UL
Comprehension	.377	.095	.143	.190	.564	<.001
Readability	−.543	.141	−.139	−.821	−.266	<.001

Research question 4

Results from linear regression show that the participants’ perceptions of researchers did have an effect on the level of comprehension of the consent form, F(1, 747) = 19.63, p < .001, R² = .03, and perceived readability, F(1, 747) = 28.74, p < .001, R = .04). Demonstrating that increased trust in researchers resulted in higher comprehension of the informed consent. See Table 7.

Table 7. Regression Model for effect of trust on comprehension and readability.

	b	SE	β	95% CI		p
				LL	UL
Comprehension	−.503	.114	−.160	−.726	−.280	<.001
Readability	.897	.167	.192	.569	1.226	<.001

Qualitative results

The two open-ended questions asked participants why they did not read the consent form if they indicated they had not and how they felt consent forms could be improved so that more survey takers would read them. See Tables 8 and 9 for details of participants’ open-ended responses. For the first question, 39.1% of participants (N = 293) indicated they read the consent form in full. Participants who said they had either skimmed or did not read the consent form gave various reasons for doing so. About a quarter of participants indicated they did not read or skimmed the form because it was too long (20.8%, n = 105) and because all consent forms looked the same (17.6%, n = 89). Participants also indicated they did not read the consent form because they felt the study (and online surveys in general) presented no risk to them personally (9.9%, n = 50). For example, one participant noted, “For a questionnaire with no identifiable risk, reading carefully is a waste of my time.” A few of the participants (2.4% n = 12) also noted that the language of consent forms typically consisted of “jargon” or “legalese,” which they did not want to read. Responses to how consent forms could be improved were similar to those of previous informed consent studies (Geier et al., 2021). The participants recommended that the best way to enhance consent forms would be to make them shorter (38.3%, n = 349) and to use simplified language (15.7%, n = 160). Some respondents also suggested visual changes to the forms, such as using bolded/colorful/highlighted text to emphasize importance (9.1%, n = 83), using bullet points rather than paragraphs (3.8%, n = 35), or putting the most critical information at the top of the form (3.5%, n = 32).

Table 8. Frequencies of participants’ coded open-ended responses to why they did not read the consent form (Q15).

Response code	n (%)	Sample quote
To long/too many words /wanted to get started	105 (20.8)	“Because it resembles the length of war and peace”
All the same	89 (17.6)	“After so many surveys they all look the same”
Answer N/A/not enough info in answer	84 (16.6)	“Not sure”
Didn’t want to/lazy/careless	66 (13.0)	“Because I didn’t want to”
Not important/not necessary /not concerned about risk	50 (9.9)	“I feel the risk was not great enough to bother”
Skimmed it/read part (but not all) of it	45 (8.9)	“I did skim for what I consider to be the critical parts.”
Couldn’t find/didn’t see/ didn’t realize it was a consent form/ nothing to do with research /not part of study	29 (5.7)	“i thought it was the initial page to the study”
Didn’t want to read legalese/jargon	12 (2.4)	“Because I’m sick of unnecessary legal verbiage that doesn’t mean anything to me”
Trust the source (e.g. institution or website administering survey)	9 (1.8)	“I feel safe doing surveys on this web site”
I did	7 (1.4)	“I did read it.”
Trouble concentrating/focusing/multitasking	6 (1.2)	“I was interrupted several times”
Required to get paid	4 (.8)	“you say no you don’t get paid”
Total	506 (100)

Note: The total N is greater than the number of participants who responded to this question as participants’ responses could correspond to multiple codes.

Table 9. Frequencies of participants’ coded open-ended responses to how consent forms could be improved (Q17).

Response code	n (%)	Sample quote
Shorter/more concise	349 (38.3)	“It should be short and straight to the point”
Answer N/A/not enough info in answer	160 (17.5)	“don’t know”
Simplified language/less jargon	143 (15.7)	“That they are in simple English and easy to read and understand”
Bold/colorful/ highlighting/bigger font	83 (9.1)	“Try to highlight important key points”
Nothing will help/no changes needed	41 (4.5)	“no way. Some will read and others just do not care”
Use bullet points/break up text/ layout changes	35 (3.8)	“Bullet points not blocks of text”
Put important info at the beginning	32 (3.5)	“Listing all important information on top”
Interactive/engaging/ask questions	30 (3.3)	“not allow people to move forward without asking questions”
Consent not needed	10 (1.1)	“no need for consent forms in the first place”
Manipulation feedback (specific to form they saw, not forms in general)	9 (1.0)	“Tell people that there will be specific questions relating to the consent form.”
Offer incentives for reading	6 (.7)	“offer bonus money if read”
Pictures	4 (.4)	“graphics”
Force time the consent /make reading required	4 (.4)	“make them so you can’t go past them without doing so”
Top header (READ, IMPORTANT INFORMATION, etc.)	3 (.3)	“emphasis at the very top as to why a reader should pay attention”
Educate people on importance of consent forms	3 (.3)	“Start up front with telling people why they need to read the consent form.”
Total	912 (100)

Note: The total N is greater than the number of participants who responded to this question as participants’ responses could correspond to multiple codes.

Largely, participant responses seemed to be based on their more general experiences with consent forms rather than their specific experiences with the version of the consent form they saw for this survey. None of the changes suggested by participants mapped significantly to the version of the consent form they were randomized to for this survey.

DISCUSSION

This study examined six variations of the same consent form to understand which version of the consent had the best comprehension. The interactive conditions (Interactive Control and Interactive 4^th Grade Reading Level) had the highest comprehension and readability scores compared to the other conditions in our study. Based on these findings, consent forms should be written at a lower grade level and have some interactive elements to help encourage individuals to read consent forms easily and understand what they are consenting to.

Our study’s findings, particularly in the comprehension in interactive consent forms, prompt a critical examination of what constitutes “informed” in informed consent. While these scores indicate a higher comprehension level than traditional methods, informed consent transcends content comprehension. It embodies a participant’s ability to not only understand the information presented but also to engage with it critically. This involves the capacity to ask questions, seek clarifications, and reflect on how the information pertains to one’s personal context. Thus, a truly informed consent process is dynamic.

Hence, while our interactive forms have demonstrated effectiveness in enhancing comprehension, we must consider if this metric alone suffices for informed consent. A participant scoring less than perfect may still be “informed” if they have had the opportunity to engage meaningfully with the content and have their concerns addressed. Conversely, a perfect score may not guarantee informed consent if it was achieved without deeper engagement or reflection.

Additionally, this study found that individuals’ perceived levels of self-efficacy influenced comprehension and readability of the consent form. While our study aligns with the observations of Perrault and Keating (2018) on improving informed consent processes in online research, it further extends the understanding by demonstrating the influence of self-efficacy and trust in researchers. Unlike Varnhagen et al. (2005) and Douglas et al. (2021), which focused primarily on the readability and comprehension of consent forms, our study introduces the additional dimensions of self-efficacy and trust, shedding light on underexplored factors that may enhance participant understanding and engagement. The relationships between self-efficacy and comprehension and readability of the informed consent were not seen previously in a student-based population study (Geier et al., 2021). However, these results align with multiple studies that suggest higher self-efficacy increases understanding, readability, and performance (Sofyan & Jayanti, 2019; Yogurtcu, 2013). The participants’ perceptions of researchers were also influential on comprehension and readability. Although no research has explicitly looked at how an individual’s trust in scientists affects the comprehension and readability of consent forms, trust in scientists is a strong predictor of attitudes and behaviors as seen within this current study’s results (Kim & Fang, 2020; Marques et al., 2015). Thus, researchers interested in improving the informed consent process should explore how building trust between researchers and research participants can improve participants’ experience with the readability and comprehension of the consent process.

Lastly, when asked how consent forms could be improved, many of our participants suggested making visual changes (e.g., bolded, coloring, and highlighting and emphasizing important information). Many responses from participants also indicated that they regularly completed a high volume of surveys and were attempting to get through them quickly to earn more money. Based on these findings, consent form writers need to know their participants’ motivations to ensure their participants feel like they are not “wasting their time” when reading the consent forms.

IMPLICATIONS FOR PRACTICE

Best practices

Researchers working with student and non-student populations need to increase their efforts to educate participants about the importance of the consent process. More importantly, researchers should understand how these populations differ in how they characterize risks. Furthermore, researchers interested in improving reading and comprehension of informed consent should consider implementing interactive features within their design. This data suggests that participants in the Interactive conditions had higher comprehension levels than participants in other non-interactive conditions. We believe this may help improve the data quality provided by the participants.

Many participants perceived low or no risk to online studies, thus feeling no reason to read the consent language. However, researchers know this is not always the case, particularly when surveying participants about sensitive topics. If researchers do not carefully and concisely provide the risks and benefits of each study and ensure that participants are reading this information, we have ethically failed to protect the participants. Thus, researchers should design better consent forms that effectively communicate risks and benefits. This will help ensure better data quality and possibly improve the trust in researchers.

Future research

Few have examined the influence of interactivity on the readability and comprehension of consent forms. Our study’s findings reveal that the interactive forms increase participants’ understanding and legibility of the consent forms. Future research should explore additional interactive features that could improve the consent process for participants, particularly with the increase in online studies. Moreover, findings from our research suggest that self-efficacy and trust in science are related to comprehension and readability. For example, participants with increased self-efficacy and trust in science have improved the readability of the consent forms. Few studies outside of this current research have explored how to improve trust in science among non-student-based populations or how self-efficacy for these populations impacts the informed consent process. Future research should seek to improve understanding how participants from various backgrounds view consent forms and trust issues with researchers and the academic research community.

Participants’ qualitative feedback in our study suggests that some feel like the informed consent process is used to solely protect the universities that administered the informed consent and not the participants. This finding also highlights how qualitative research may enhance our understanding of how participants feel about the informed consent process in ways quantitative-based measures cannot. Thus, we also propose that more qualitative research is needed to understand the barriers to consent from comprehension and the comprehension of risk. Understanding how these populations define the risks and benefits of various research topics (e.g., sensitive versus nonsensitive) to help design informed consents tailored to these perceptions.

Limitations

As with every study, some limitations need to be considered. Although the panel service we used attempted to match general U.S. demographics when recruiting respondents, our sample population was more white, male, and educated than the general U.S. population. This matches general trends in online survey research using non-probability sampling (Lehdonvirta et al., 2021). Additionally, we recruited participants from a panel service that included participants familiar with taking research limitations studies. We did not specifically ask how much experience the participants had with online surveys, nor did we consider the effect of their experience. There may have been some habituation effect. However, most online surveys utilize a panel service; thus, our study still provides vital information about how a general online survey perceives consent forms for online studies. Last, as participants were informed the study examined informed consent, they may have been primed to read the forms more thoroughly than in real-world settings where consent forms cover diverse topics. Future studies could mitigate this limitation by assessing comprehension of consent forms without revealing the study’s purpose to participants.

CONCLUSION

Findings from this study shed light on important considerations for improving the consent process, particularly for general populations, by identifying potential barriers to participants’ reading and comprehension of informed consent forms. Our study revealed that participants expressed dissatisfaction with the conventional nature of most consent forms and indicated that their perception of study risk influenced their engagement with and understanding of the forms. Notably, an interactive consent form at a lower reading level demonstrated a positive impact on participants’ comprehension of the study’s requirements. Based on these findings, we recommend that researchers take proactive steps to enhance participant engagement using interactive consent forms, particularly for studies with low levels of risk. Such forms should aim to provide concise and easily comprehensible information at an appropriate reading level. By adopting this approach, researchers can foster a more meaningful and informative consent process that respects participants’ autonomy and ensures their informed decision-making.

It is crucial for researchers to recognize that consent forms should not be treated as mere formalities but rather as an ethical commitment to genuinely inform participants about the nature and expectations of their involvement in the study. By valuing the consent process as an ethical obligation, researchers can contribute to a more inclusive and participant-centered research environment.

Lastly, this study highlights the significance of addressing participants’ concerns and preferences when designing consent forms. By embracing interactive and accessible approaches to consent, researchers can build trust, promote comprehension, and empower participants to make informed decisions about their involvement in social sciences research. Continued efforts to improve the consent process will lead to enhanced participant experiences and strengthen the ethical foundations of scientific inquiry.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

DATA AVAILABILITY STATEMENT

Data available on request from the authors.