Testing the impact of culturally-relevant communication style on engagement with Hispanic and Latinx adults

Abstract Effective science communication for a multilingual population requires more than language translation, it also requires being mindful of cultural communication styles. This study tested the impact that communication style has on feelings of inclusion, learning, and engagement in the Earth sciences for Hispanic and Latinx adults. An online survey with open and closed questions was used to evaluate two science videos (in both Spanish and English) with different communication styles: 1) a traditional, interview-based style, where experts present a science concept, and 2) an informal conversational style, where a scientific message is shared through a casual conversation. Seventy-four participants self-identified as Hispanic and Latinx and were the focus of the data analysis. Both video styles were positively received, with participant feedback emphasizing feelings of inclusion in seeing Latina scientists, easy to understand science concepts, and accessible language. Hispanic and Latinx adults preferred the traditional video, but the conversational video ranked higher in other aspects, which varied based on participants’ primary spoken language at home. For example, the conversational video had a positive impact on the ability to relate information to their own life and increase awareness of Earth science careers for Spanish-language speakers. Findings suggest the use of both video styles could improve feelings of inclusion and engagement for Hispanic and Latinx adults. Additional aspects of culture and demographics may explain some of the language-based results. Future science videos are encouraged to be co-designed by, for, and with Hispanic and Latinx communities to emphasize cultural values while avoiding stereotyping and cultural appropriation.


Introduction
Although English proficiency has increased and Spanish use at home has decreased in the United States (U.S.) since 1980, especially for U.S.-born Hispanics and Latinx, 70% of total Hispanics and Latinx in the U.S. speak only Spanish at home (Krogstrad & Noe-Bustamante et al., 2021). Scientific literacy in geosciences, particularly as it pertains to climate change and extreme weather events, is critically important for vulnerable populations, such as Hispanic and Latinx communities in the U.S. (Akerlof et al., 2015;Benevolenza & DeRigne, 2019;EPA, 2021;Yas et al., 2021;Trujillo-Falcon et al., 2021). Despite rampant misinformation pertaining to climate change (Treen et al., 2020), studies show that Hispanic and Latinx people are more likely to agree with scientific evidence that climate change is anthropogenic in nature and that environmental issues impact their local communities (Krogstad, 2015;. Creating content that is culturally appropriate and regionally specific to these populations can build trust in scientists, save lives, and increase the sense of belonging in the Earth sciences. A large body of research and community of practice exists for engaging with the public on geoscience topics (e.g., Liverman et al. (2008) and references within), but the scientific content is limited for Spanish-language speakers in the U.S., hindering educational reach to Hispanic and Latinx communities (Medina more indirect, passive communication style, providing high context, with use of gestures and minimal eye contact (Kaplan, 1967;Hall, 1976;Cruz, 2001;Lewis, 2006;Catalaa, 2019a). Additionally, Hispanic and Latinx people have an oral tradition of passing down knowledge through stories and common sayings (or dichos in Spanish), with family and community being a big part of their identity (Sánchez et al., 2010;Reese, 2012;Torres et al., 2016;Mizrahi, 2019). Therefore, when creating science videos in Spanish in the U.S., it is not just about translating the language from English to Spanish, but rather it is about ensuring that both communication styles and cultural values (SciGirls, 2010) also translate to reach the community of interest. A lack of communication style translation may result in misunderstanding the message, perceived biases, and challenges with information retention for Hispanic and Latinx communities (Cruz, 2001;Lewis, 2006;Torres et al., 2016).
A large share of the U.S. population is composed of Hispanic and Latinx people, yet they continue to lack representation in STEM careers (Bernard & Cooperdock, 2018;U.S. Census Bureau, 2021;Morales et al., 2022). Research has shown that parents and other adult members of Hispanic and Latinx families have a comparatively large influence on children's career choices and their engagement in school (Chrispeels & Rivero, regardless of race. The term Latinx is the gender-neutral alternative of Latino or Latina, thus "Hispanic and Latinx" is used here as an inclusive term to refer to this ethnic and cultural demographic group. Most Hispanic and Latinx adults in the U.S. are not familiar with nor use the term Latinx , thus the mindful decision was made to ask participants if they were of Hispanic, Latino, or Spanish (HLS) origin within the survey module. Within the manuscript, "HLS" or "HLS origin" will be used to present the survey module data and may be interchangeably used with "Hispanic and Latinx" throughout.
The mission of the National Center for Atmospheric Research (NCAR) is to support knowledge transfer and extend capabilities of the broader national and international scientific community. NCAR is working to equitably engage with minoritized communities in STEM, including communities of HLS origin, to increase interest in STEM careers and scientific literacy (NCAR, 2019). The authors currently engage Spanish-language speaking audiences through translations of educational and outreach resources (e.g., museum displays and career-focused resources), as well as in-person and virtual Spanish-language opportunities (e.g., Medina Luna & Zietlow, 2019;Medina Luna et al., 2021).
Within the scope of this study, accessibility means communicating science concepts in a language (Spanish or English) that an HLS audience can understand (i.e., commonly used vernacular) with scientific explanations provided at an introductory level for the general public, and with few barriers to access the information (e.g., no cost, available any time online, captions provided, short duration). Accessible scientific content can facilitate inclusion, which corresponds to a viewer feeling connected with and represented by the characters and/or topic presented in the videos. Interest and engagement correspond to piquing a viewer's interest to seek additional information, wanting to share the information with friends and family, having the material keep their attention, and feeling like they can be part of the conversation. While engagement of an adult HLS audience was the primary motivation, understanding if a conversational video style in either language could facilitate learning was also of interest.
The survey was designed to assess the video communication style preference of HLS adults, as well as the impact that video style has on learning, inclusion, and engagement with this community (Figure 1). An online survey module was designed with closed questions to gather information on preference and rankings for each video style, and open-ended questions to gather written responses to understand the video style rankings and preferences selected by participants. The open-ended questions also provided a space for participants to provide general feedback on the videos and advice for creating inclusive science videos in the future.
When collecting data from Hispanic and Latinx communities, it is recommended to use group dialogue and reflection, to partner with organizations already working with the community, and to build trust between the researchers and community of interest (SciGirls, 2010;Wills, 2012;Marcum, 2018). Unfortunately, the COVID-19 pandemic restricted data collection to an online platform, a digital survey. Information about participating in this study was distributed through social media, email listservs, and physical flyers printed and placed in local businesses. The online survey approach provided an effective way for participants to compare the video styles being tested, and included both domestic and international participation from HLS communities.

Video creation and description
A total of four videos were created, one for each communication style tested, and versions of each video were also made in English and Spanish. This study opted to create virtual media as a way to increase accessibility and reach broader audiences (e.g., Forrester, 2020). The videos were created in-house using the video production expertise of the coauthors. The videos were created within a one-month time period, which included scriptwriting, production planning (e.g., location scouting, developing shot lists, obtaining proper permissions), filming, editing, and reviewing ( Figure 2). The Spanish and English language scripts were developed separately, but in tandem; the Spanish language scripts were not simply translations of the English video scripts. Two native Spanish speakers reviewed the scripts for accuracy and cultural relevance, and were paid for their time for this review. Due to COVID-19 safety guidelines, the number of people that could be on-screen simultaneously was limited, and filming was completed outdoors over two separate afternoons, to limit virus transmission and to ensure proper lighting and good audio quality. Editing of the videos was done using Adobe Premiere Pro, with Adobe Audition and Adobe After Effects was used for audio editing and creation of the animations. After completion of the videos, two native Spanish speakers reviewed the videos and feedback provided was implemented.
The videos aimed to share the following overarching message: forecasting precipitation over mountains is challenging because mountains influence local weather. The topic was selected to acknowledge that, as described in Section 1, Hispanic and Latinx people in the U.S. are at a high risk of being exposed to weather and climate hazards (EPA, 2021;Yas et al., 2021;Trujillo-Falcon et al., 2021), and  emphasizes the need to reach this community with more than a language translation. Additionally, the work of author A. Morales is associated with understanding cloud and precipitation processes over mountains using numerical weather models (e.g., Morales et al., 2015).
The videos feature two women of Mexican-American heritage speaking in a dialect corresponding to the researchers' (Morales and Medina Luna's) heritage. The dialect on-screen reflects the nearly 62% of the U.S. HLS population who are people of Mexican origin (Noe-Bustamante, 2019). Two native Spanish-language speaking atmospheric scientists with Puerto Rican and Nicaraguan heritage reviewed the video scripts to ensure the videos were accessible to Spanish-language speakers with other cultural dialects.
Each video is about three minutes or less ( Figure 2) in length to avoid a lapse in attention or interest while watching the videos and completing the survey module. The traditional and conversational videos in Spanish were 26 and 9 seconds longer than the English versions, since conveying a similar message in Spanish often requires additional words and phrases. A considerable difference between the traditional and conversational video styles is the amount of b-roll, or relevant visuals other than a person speaking on-screen, used. The traditional and conversational videos have ∼85% and ∼25%, respectively, of their video length with b-roll (i.e., time when the speakers were not shown on screen). The URLs to each video can be found in Figure 2 and are free to share.
The control video style (hereafter termed "traditional") uses a formal, interview-based format to communicate science. The traditional style videos are similar to those commonly produced by NCAR, and other scientific organizations, to share interesting research and field work to the general public (e.g., Medina Luna & Zietlow, 2019). Specifically, a narrator and a field expert speak to the audience and visual aids (e.g., animations of science concepts) are shown throughout the video ( Figure 2). The science expert is shown "at work" and the video ends with information on how viewers can help scientists obtain additional information about weather forecasts. The traditional format videos are titled, "The Complexities of Forecasting Precipitation Over the Mountains/Las Complejidades de Pronosticar la Precipitación Sobre Montañas." The experimental video style (hereafter termed "conversational") uses a comparatively less-formal format with two friends having a conversation about the weather in an outdoor picnic setting with mountains in the background ( Figure 2). Although one friend casually states that the other is a scientist, the expertise and educational credentials of the characters/actors are not shown until the end of the video. The conversational videos are titled "All Weather Friends/El Tiempo Entre Amigas", which is a play on words in Spanish because time and weather are both translated as "el tiempo." The conversational video, as the title describes, focuses on a conversation between friends and thus offers fewer visual aids to describe science concepts, but rather focuses on communication styles traditional of Hispanic and Latinx cultures (e.g., gestures, tone of voice). The video ends with a common saying or dicho, "… it's better to be safe than sorry/así que más vale prevenir que lamentar." The saying was added as suggested by Sánchez et al. (2010) that the use of dichos is a culturally-appropriate manner of facilitating communication within HLS Spanish-language speaking families.
The motivation for the conversational video included both cultural and practical reasons. The conversational video style was inspired by telenovelas, which are Spanish language soap operas that can help HLS immigrants remain connected to their cultural roots (e.g., Frazier et al., 2012). Thus, the conversational video has upbeat music and a title card at the beginning of the video, after which the two characters begin interacting as they set up their picnic. When engaging with HLS communities, it is recommended to incorporate cultural values into programs/materials, e.g., family, youth, cultural art, food and music (SciGirls, 2010;Wills, 2012;Marcum, 2018).

Online survey design
The online survey module was designed in SurveyMonkey (Abd Halim et al., 2018). Survey questions were developed by first defining the project goals to focus the evaluation, as well as the study population. As presented earlier, the survey sought to evaluate the impact of video style on the themes of inclusion, learning, and engagement. Spanish-and English-language versions of the survey were developed to reach a broad audience of HLS adults. To ensure accurate interpretation of the findings, survey items were reviewed by the researchers for face validity. The Spanish version of the survey module was reviewed and refined by a professional science translator who understood colloquial use of terminology and phrasing that would be best understood by study participants. The Spanish survey was also reviewed by the Spanish-language speakers who had reviewed the video script. The English version was reviewed by English-language speakers as well. The survey allowed an opportunity for participants to provide guidance and suggestions on how to improve the video series to make future videos culturally inclusive. Thus, the survey included both closed and open questions. The survey questions were submitted to and approved by the NCAR internal review board. Design of demographic questions was guided by NCAR's Office of Diversity, Equity, and Inclusion.
Participants chose whether to take the English or Spanish survey, which then determined the language of the videos they would watch. The survey module contained an introductory overview of the project and an informed consent slide. Participants were then asked to answer eight questions to assess their experience and knowledge with scientific concepts prior to watching the two video styles. Subsequently, participants viewed one of the two videos, answered a series of questions related to the first video, and then completed the same process for the second video. The order of the videos was randomized in each module to control for changes or bias in response based on viewing order.
Participants were required to identify the name of the video they viewed first and second, prior to answering questions about each video. For each video, participants were asked to respond to the same three-to-four questions/statements in each of the following categories: a) knowledge/learning, b) video (composition and design), c) inclusivity, and d) interest/engagement (see Supplemental Files for the survey). Participants were also asked to compare the two videos and indicate their preferences in terms of various statements related to the key themes of the research. The module then concluded with demographic questions to better understand the participants' background and to ensure future implementation adequately addresses different viewer needs.

Data collection
The goal for data collection was to reach a wide array of participants, including a mix of adult Spanish-, English-, and bilingual speakers from HLS and non-HLS backgrounds within the U.S., and gather as many responses as possible. Survey module links and QR codes were generated in SurveyMonkey and distributed by the researchers via email mailing lists and community boards (in person and online). These were distributed to professional societies (e.g., American Meteorological Society, American Geophysical Union, Society of Latinx/Hispanics in Earth and Space Science), alumni networks (e.g., UCAR's Significant Opportunities in Atmospheric Research and Science), and local community groups (e.g., El Centro Amistad, Casa de la Esperanza Learning Center). The survey links were also distributed throughout social media platforms (e.g., Facebook, Twitter, Instagram) and flyers were posted on community boards or in local businesses owned and/or patronized by Hispanic and Latinx adults in Boulder, Colorado, the geographic location of the authors of this article. The survey was also shared via word-of-mouth to family, friends, and on personal social media accounts in order to reach participants who are not affiliated with the weather, water, and climate enterprise. The survey was made available for two weeks, September 8-21, with a reminder sent out on September 15, 2021 to coincide with the start of Hispanic Heritage Month in the U.S.
It is unknown how many individuals were reached through the distribution methods, but a total of 186 surveys were completed; 94 individuals completed the Spanish-language version and 92 the English-language version. As this study focuses on the impact video style has on Hispanic and Latinx adults, the data and results presented herein represent only the 74 participants who self-identified as being of HLS origin.

Data analysis, limitations and delimitations
The present study is delimited in nature by its intent to gather feedback and serve as a foundation for future studies. As such, the study focuses on descriptive analyses and qualitative feedback to answer important questions related to the ways in which different audiences learn about weather and weather forecasting. Given this focus, the online survey data were analyzed using thematic and descriptive analyses to provide information on participants and their preferences for video style as well as the impact of the videos on participants' understanding of the concepts described in the videos. To this end, the researchers did not conduct validity and reliability testing of the survey (beyond face validity). Therefore, the results of the present study need to be interpreted with caution, and additional testing needs to be conducted before the survey can be generalized or scaled up for future use. Additionally, given the small sample size, the results need to be interpreted with further caution.
A limitation of the present study is that participants selected whether the survey module they completed was in English or Spanish. Thus, researchers had no control over the video series that would be viewed or the representativeness of the sample. Additionally, the intent was to gather feedback from the U.S.-based HLS general public; however, online distribution of the survey module links was not constrained by geographic location. Despite these limitations, the survey sample was fairly representative and a similar number of responses were obtained for both the Spanish and English versions of the survey modules.

Study population
To better understand the differences in video style preference for Hispanic and Latinx adults, the datasets from the English-and Spanish-language surveys were combined and analyzed for themes with the 74 participants of HLS origin ( Figure 3). Of these 74 participants, 53 opted to take the survey in Spanish, and 21 took the survey in English, which also determined the language of the videos that they watched. Note that selection of survey and video language does not correlate to the language primarily spoken at home that was selected by participants.
HLS participants were primarily within the age range of 25-44, with a larger percentage of participants identifying as a woman for ages 25+ (Figure 3b). Almost half of HLS participants grew up in a mountainous area (Figure 4). Most thought that mountains can affect the weather and weather forecasting (72%), were interested in weather forecasting (66%), had watched a documentary, television show, or video about weather forecasting (54%), and indicated the weather affects their day-to-day activities (65%, Figure 4). A small percentage of HLS participants were interested in (23%) or had a career in weather forecasting or computer modeling for weather prediction (19%). Over two-thirds (68%) had taken a class on weather forecasting (Figure 4).
Participants self-selected the language they most often spoke at home and of the 74 HLS participants, 40 (54%) primarily spoke Spanish at home, 14 (19%) primarily spoke English at home, and 20 (27%) were bilingual in Spanish and English (Figure 3a). A high percentage of Spanish-language, English-language, and bilingual speaking HLS participants indicated they thought mountains can affect the weather and weather forecasting (70%, 79%, 70%, respectively), were interested in weather forecasting (68%, 71%, 60%, respectively), and indicated the weather affects their day-to day activities (63%, 71%, 65%, respectively). Interestingly, at least half of Spanish-language speaking (50%) and bilingual HLS participants (55%) grew up in a mountainous area compared to 21% of HLS English-language speakers. In contrast, a higher percentage of HLS English-language speakers indicated they had taken a class on weather forecasting (64%), had watched a documentary, television show, or video about weather forecasting (86%), or had learned about weather forecasting from the weather segment on the news (71%). These results were much lower for HLS Spanish-language and bilingual speakers: had taken a class on weather forecasting (30% for each), watched a documentary, television show, or video about weather forecasting (50% and 40%, respectively), learned about weather forecasting from the weather segment on the news (40% and 45%, respectively). Less than a third of Spanish-language, English-language, and bilingual speaking HLS participants indicated they had a career in (25%, 7%, 15%, respectively) or were interested in pursuing a career in (30%, 21%, 10%, respectively) weather forecasting or computer modeling for weather prediction.

Results
Both video styles were well-received, with participants sharing positive feedback that the science information was easy to understand and they enjoyed seeing Latinas in STEM. That said, 51% of all HLS participants preferred the traditional video style compared to 29% who preferred the conversational style and 14% that chose both styles ( Figure 5).
Based on written feedback, the traditional style was preferred because the video 1) was more direct and had detailed science content, 2) showed scientists at work, 3) had more visual aids, 4) shared options to take action (e.g., community or crowd-sourced science), and 5) the more professional Figure 3. study population demographics. Panels show the distribution of a) languages most often spoken at home, self-selected by participants who are of Hispanic, latino, or spanish (Hls) origin, and b) their age groups and gender distribution (full bar is the % of total Hls participants [n = 74], sub-bars are the gender percentages within each age group, as indicated in the legend). adding all age group percentage categories equals 100%.
format had "better acting. " One specific comment mentioned the traditional video had a higher level of complexity while remaining clear and simple, "tiene un nivel de complejidad un poco más alto que el otro, sin dejar de ser sencillo y claro. " Another participant noted that although the conversational video had a similar message, it was too informal and looked like a telenovela (i.e., soap opera).
For those who preferred the conversational style, the main reasons provided were that the video 1) had a story, information, and manner of speech that were more relatable, 2) was easier to understand, 3) was more fun, and 4) provided a more casual learning environment. One participant noted that the conversational video provided educational material "in a casual, relaxed setting among friends where you don't realized [sic] you are learning", while another participant stated that the traditional video "felt like a science video from my middle school days." Participants who had no specific preference on the video style and selected "both videos" stated both videos were interesting, had easily understandable language, and presented a topic related to their experience living in the mountains. One participant encouraged the use of both videos in tandem to set up an approachable introduction before providing more detailed explanations, "I think both videos are important and need to go hand in hand, it is important to see yourself represented and to have a regular conversation that incites curiosity. I watched [the conversational video] first which piqued my interest and then watched the [traditional video] about forecasting which continued to explain what I became curious about in the [conversational video]. I appreciate that I had an intro that was welcoming and then a more scientific approach that was also very easily digestible. " Based on specific questions assessing learning, inclusion, and engagement, the traditional video style ranked higher in nearly . Percentage of all Hls origin participants (n = 74) which selected that the statements listed applied to them, asked before the videos were shown to assess prior knowledge and experience.

Figure 5.
Video style preference for all Hls origin participants (percentages are based on the total sample size of n = 74). the sub-bars show the percentage of Hls origin participants' video preference based on primary language spoken at home (as indicated in the legend). note: additional options such as "neither", "i don't know", and "no response" are not included in this figure; these other responses correspond to 7% of Hls origin participants (n = 5). all aspects. Most notably, 71% of HLS participants agreed or strongly agreed that the traditional video made them aware of career options in the Earth sciences (Figure 6), 12% higher than after watching the conversational video. However, 93% of HLS participants agreed or strongly agreed that they could relate the science information presented to their own life after watching the conversational video, 7% higher than after watching the traditional video ( Figure 6). Additionally, 5% more HLS participants understood the overall science message to be that predicting the weather is more difficult over mountains after watching the conversational video (Table 1).

Video style preference and impact based on primary language spoken at home
Preference for the traditional video style was consistent regardless of language most often spoken at home. HLS participants who primarily spoke Spanish at home had similar preferences between traditional and conversational styles (22% and 19%, respectively), implying a less definitive video style preference compared to bilingual and English speakers ( Figure 5). Although the traditional video did rank higher with respect to assessed aspects of learning, inclusion, and engagement, there were a few notable exceptions.

Spanish-language speakers
Interestingly, 90% of Spanish-speaking HLS participants said the conversational video made them feel like they were being taught a scientific concept, 9% more than after watching the traditional video (Table 2). Although nearly all Spanish-speakers thought both videos provided scientific information at a level they could understand, 5% more felt the conversational video provided more understandable material (Table 3). All Spanish-speaking HLS participants agreed or strongly agreed that the conversational video presented science information they could relate to their own life, 12% more than after watching the traditional video (Table 4). Survey feedback indicated that Spanish-speaking participants liked the conversational video style because they thought it was more fun and engaging, made them feel more included, and had a communication style that was more relatable. Presenting science information that is relatable and culturally relevant for the HLS origin demographic was one of the goals met by this study.

English-language speakers
Although more English-speaking HLS participants agreed or strongly agreed that they learned something new after watching the traditional video (Table 2), 14% more agreed or strongly agreed that they were able to learn something new in a way that made them feel included and like they could Figure 6. results of video style impact on aspects of learning, inclusion, and engagement for all Hls participants (n = 74). Bar chart shows the percentage of all Hls participants who agreed or strongly agreed with each statement above the bars after watching the traditional (blue) or conversational (yellow) video style. Bold text is provided to more easily distinguish between statements; it has no significance to results. Percentages of all Hls participant responses and Hls participant responses broken up by primary language spoken at home. note: additional response options were "easier" or "i don't know," which make up the rest of the 100% total for each group.
be part of the conversation after watching the conversational video (Table 4).

Bilingual speakers
After viewing the conversational video, 19% more bilingual HLS participants learned the desired scientific concept compared to those who watched the traditional video (Table 1). Five percent more bilingual HLS participants felt comfortable sharing what they learned with peers, family or friends after watching the conversational video compared to the traditional video (Table 2). Likewise, 5% more bilingual HLS participants agreed or strongly agreed that the people in the conversational video spoke a similar dialect as them compared to the traditional video (Table 5). In the same vein, when asked if the words and phrases in the video made sense to bilingual viewers, 5% more agreed or strongly agreed with the statement after watching the conversational video, compared to the traditional video (Table 3).

Conversational video impact based on primary language spoken at home
Analyzing the results for the conversational video only, there were a few striking differences across languages spoken. A larger percent (82%) of Spanish-language speakers felt they learned something new about weather forecasting that they did not know before compared to bilingual (64%) or English-language speakers (60%; Table 2). Watching the conversational video style made 71% of Spanish-language speakers aware of career options in the Earth sciences. That is, at least 21% more than for the English-language speakers (50%) and bilingual participants (40%; Table 6). The conversational video had a greater positive impact for English-language speakers than bilingual or Spanish-language speakers, with respect to feeling comfortable asking questions or seeking additional information (93% ; Table 4), feeling connected with the people in the video (93% ; Table 5), and making them curious about Earth sciences or weather forecasting (85%; Table 6).

Notable differences based on viewing order
Results indicate that HLS participants were able to identify the science message more easily in the conversational video, only after watching the traditional video first. However, given the small sample sizes in some cases (e.g., 3-4 people), this suggested finding would need to be explored in more detail in future work. More HLS participants, regardless of language spoken, who viewed the traditional video first agreed rather than strongly agreed that they could relate to the science information presented (N = 15 agreed, 12 strongly agreed) and that they learned something new in a way that made them feel included and like they can be part of the conversation (N = 16 agreed, 12 strongly agreed).
More HLS participants who first viewed the traditional video agreed (N = 18) compared to strongly agreed (N = 10) that the people or the narrator spoke the same dialect as them (not shown in tables); whereas more HLS participants who viewed the traditional video second (after first viewing the conversational video) strongly agreed (N = 24) compared to agreed (N = 14). A greater number of Spanish-speaking respondents who viewed the traditional video first agreed rather than strongly agreed that the people and the narrator spoke their same dialect; this was also true for English-speaking respondents. Percentages of Hls participant responses based on primary language spoken at home who agreed or strongly agreed with each italicized statement after watching the traditional or conversational video style are provided. The number of HLS participants who viewed the conversational video first and who agreed with the statement, "I felt connected to the people in this video," was slightly higher than the number of HLS participants who strongly agreed with the same statement (N = 18 agreed, 17 strongly agreed). However, a greater number of HLS participants who viewed the conversational video second strongly agreed (N = 17), compared to the number who agreed (N = 10).

Participant feedback on the video series
HLS participants offered several recommendations for what could be done differently to make the videos more inclusive, engaging, and educational. Both videos had an overall positive response, with many participants commending the researchers on a job well done and wanting to see more of both types of videos. Several members of the HLS community mentioned feeling represented or included in the videos, particularly because the presenters were Latina. Several participants recommended more clearly designating the ideal audience for each video. For example, many felt that the language and content in the conversational video style was at an introductory level, but that may have been influenced by the viewer's prior knowledge, learning, or experience with weather and weather forecasting over mountainous areas.
Some feedback provided suggestions toward improving the conversational video style. Viewers suggested improving the production quality (e.g., more fluid speech) and incorporating more animations or cartoons. Having a group conversation instead of a one-on-one chat could make the format more dynamic and give the opportunity to represent more of the cultural and ethnic diversity that exists within the HLS community.
Participants were interested in videos set in tropical or subtropical regions, big cities, coastal regions, and seeing real weather happening, especially the type of meteorological phenomena being discussed, to add drama and excitement. Participants suggested that future videos should i) represent the diversity in Spanish dialects within Latin America, South America, and the Caribbean, ii) be fully bilingual videos, iii) be available in Portuguese in addition to Spanish, iv) incorporate American or Mexican sign language, v) include Black Hispanic and Latinx people, vi) show male and female Hispanic and Latinx scientists working together, and vii) include scientists from different latitudes, particularly those from equatorial and subtropical regions.
The traditional video style explicitly pointed out the GLOBE app, CoCoRaHS network, and the U.S. National Weather Service website to participate in gathering meteorological observations and learn more. It was suggested that an explanation or demonstration on how to use these resources would be useful to encourage participation.

Discussion
Findings from this study suggest that being mindful of communication style can positively impact learning, inclusion, and engagement with Hispanic and Latinx adults. Additional cultural aspects may also play a role. For example, Hispanic and Latinx culture values formality and deference to authority, especially in educational and professional settings, and expect educators to be experts with all the answers (Osland, De Franco, & Osland, 1999;Albert & Ha, 2004;Catalaa, 2019b). Given some of the participant feedback, perhaps Hispanic and Latinx participants associated the conversational video with some of the less positive qualities of a soap opera, such as inauthentic, dramatic, or inferior. Although telenovelas are a popular form of entertainment in Latin American homes, this may not be an appropriate format for communicating science to all HLS adults. Adult Hispanics and Latinx may associate the more formal format of the traditional video with an authoritative voice in STEM fields, thus providing more credibility and trust in the information being communicated.
Additional demographics may play a role in learning and engagement for Hispanic and Latinx adults, such as educational background, income inequality, age, or trust in authority figures (e.g., Plutzer, 2013;Barton et al., 2014;American Academy of Arts & Sciences, 2018), and may explain the language-based differences. At least half of Spanish-language and bilingual speakers grew up in a mountainous area and many of the English-language speaking participants indicated they had taken a class in, watched digital media or learned through the weather segment on the news about weather forecasting. Thus, their prior experience living near mountains or having been exposed to the science content before could have biased the educational assessment between video styles. Results for Spanish-language speakers showed they agreed that they learned something new about weather forecasting through the conversational video (Table 2), yet they misunderstood the intended scientific message after watching the conversational video (Table 1). In contrast, bilingual and English-language speakers did not feel they learned something new after watching the conversational video (Table 2), but were able to understand the desired scientific message (Table 1). Future work could investigate how prior knowledge of a topic or previous experiences impact participant's feeling of learning something new.
Building trust with the Hispanic and Latinx community, who is sometimes vulnerable to severe weather impacts, is important for effective engagement (SciGirls, 2010). Broadcast meteorologists in the U.S. can face distrust from viewers, with some receiving comments such as, "meteorologists get paid to be wrong half the time" (personal communication). One bilingual broadcast meteorologist noted hearing the phrase "mentirologos," a play on words between the Spanish word for liar (mentiroso/a) and meteorologist (meteorologo). Negative experiences or societal perceptions of the vocation of a science communicator could impact the trust a viewer has on the information they provide, potentially causing a barrier to learning and engagement (e.g., Barton et al., 2014). In this study, authors were therefore intentional about the title used for the field expert in the videos as a "cientifica/scientist" and not a meteorologist. Survey distribution emphasized that the data were being gathered as part of a postdoctoral research project aimed to improve how scientists communicate with the Hispanic and Latinx community. These actions emphasize cultural values held by the Hispanic and Latinx culture, i.e., a focus on youth, education, and helping the community (SciGirls, 2010;Wills, 2012;Marcum, 2018). The collectivistic nature of Hispanic and Latinx culture (Hall, 1976) may supersede any vocational bias and assist with scientific engagement by seeing a member of their cultural or ethnic community as the expert. Further exploration on how Hispanic and Latinx communities engage with scientists would offer opportunities to build trust and foster relationships between this community and the scientific community.

Implications on engagement
This study demonstrates how a culturally relevant science video can be created to engage with Hispanic and Latinx adults to create a sense of inclusion and access to the Earth sciences (Figure 7). Both videos were enjoyed by most participants and provided scientific information at a level most participants could understand. Using a common language, providing relatable science that a viewer could apply to their lives, and showing Latinas in STEM were key for engagement with adult Hispanic and Latinx. Survey results indicate that although Hispanic and Latinx adults preferred a more formal communication style, the more informal conversational style positively impacted other aspects of learning, inclusion, and engagement. These impacts were found to vary based on the primary language spoken at home by the adult Hispanic and Latinx participants.
Results indicate that viewing order of the conversational and traditional videos had an impact on the participant responses. Spanish-and English-language speaking participants indicated that watching the conversational video first helped pique their interest, while bilingual respondents felt more engaged with the videos after watching the traditional video which provided more detailed scientific explanations.
Results also indicate that Hispanic and Latinx participants connected more with the people in the conversational video after watching the traditional video, where the scientists proved their credibility by formally describing the science concept. Written participant feedback, which showed preference for the conversational video, described that the casual and informal format of the video made it fun to learn as "the other video [traditional video style] felt like a science video from my middle school days. " The quoted feedback is inferred as a reference to a negative experience in learning science concepts at an early age. Thus, the traditional and conversational video styles could be used in tandem to positively impact learning, engagement, interest, and inclusivity.
Notably, the conversational video style positively impacted both the relatability of the science information and the awareness of careers in the Earth sciences for Spanish-language speakers. As stated in section 1, Hispanic and Latinx adults can have a great influence on the careers chosen by children in their family (Chrispeels & Rivero, 2001;Mitchell et al., 2008;Quiñones & Marquez Kiyama, 2014;Davis & Maximilian, 2017;Milner-Bolotin & Marotto, 2018). Having Hispanic and Latinx adults understand why weather forecasting and Earth science is important to their community and everyday lives, as well demonstrating that being an atmospheric scientist or meteorologist are viable careers, may help with encouraging more Hispanic and Latinx students into STEM careers.

Recommendations
Survey feedback emphasized the continued use of easy-to-understand language and common examples to explain science Figure 7. summary of the main findings from the online survey. sP represents spanish-language speaker results, en represents english-language speaker results, and Bil represents the results of the bilingual participants.
concepts. Latino culture values connection, thus a culturally appropriate communication style can assist with inclusion and engagement in that perspective (Osland, De Franco, & Osland, 1999). Recent efforts to communicate COVID-19 health protocols in Puerto Rico showed that using culturally-appropriate metaphors was very effective (CienciaPR, 2021). In future iterations of the video series, it may be useful to increase the level of complexity in the conversational video style and potentially use both videos to describe the weather and forecasting in mountainous regions. For example, the conversational video emphasized that mountains can influence the weather, but did not provide detailed examples as to how. Adding some of the animations that were used in the traditional video could help resolve this issue. Future science videos in this series for HLS adults should include impacts of climate change around the world, highlight scientists from Latin America and Global South, use culturally relevant settings, explain weather that would disproportionately affect a particular demographic, and provide useful weather tips (e.g., identifying whether it will rain based on cloud type and color).
Future studies should understand and assess Spanish dialects in geoscience communication. For example, although the correct translation for weather is "el tiempo," the term "el clima" is a popular translation and its use varies by country of origin (Martinez & Renta, 2021). Trujillo-Falcon et al. (2021) showed that culturally appropriate translation to Spanish could help improve English terminology, thus investigating inclusivity and accessibility for one ethnic group can provide insights to improve the general communication of science. Additionally, the informality of Anglo-American culture in professional or educational settings, e.g., being on a first-name basis with teachers (Albert & Ha, 2004), suggests a conversational video style could positively impact learning and engagement with Anglo Americans. Assessing the perception of other ethnicities to a conversational video style is encouraged.
Co-producing science videos for and with the community of interest is of great importance to address complex socio-environmental challenges that impact those communities (e.g., Norström et al., 2020). Co-production can have many definitions, but typically involves a variety of different groups (e.g., western-trained scientists, community members, other stakeholders who operationalize scientific findings) working together to define a project that is actionable to community needs. It is also recognized as a more inclusive approach to project design (e.g., Teeters et al., 2016). As such, further research might include working in partnership with various HLS communities to gain insight as to what communication styles are most effective for the HLS community and also develop science stories that will be most impactful for their community.
Findings from this study indicate that there is an interest and application for culturally relevant science videos formatted in a more informal, conversational manner. Due to the variety in dialects across Hispanic and Latinx Spanish-language speakers, the language of science could be leveraged and used as a kind of neutral dialect inclusive of a diverse community. Given the interest in learning about weather and climate topics by Hispanic and Latinx adults, a great opportunity exists to learn more about how this population likes to engage in science, to build collaborations toward co-creating science materials that will be useful for their communities, and increase awareness and interest in careers in the Earth Sciences.