Williams et al., 2011). Nonetheless, Gardeners’ Perceptions of Northwestern U.S. gardens with higher proportions of na- Native Plants Are Influenced by Ecological tive plants may experience increased bee activity (Fukase and Simons, 2016) and Information and Garden Group Affiliation those with more diverse native plant groups increased pollinator diversity 1 1 1 (Hostetler and McIntyre, 2001). Aaron G. Anderson , Isabella Messer , and Gail A. Langellotto Therefore, the use of native plant spe- cies in garden habitats may help con- ADDITIONAL INDEX WORDS. gardener preferences, native plant gardening, serve broad insect groups in urban and pollinator plants, urban habitat suburban landscapes, which is particu- larly timely because of recent reports of SUMMARY. Plantings of native flowers are often installed to increase the pollinator global declines in insect populations habitat in urban and suburban gardens. However, in many regions, it is not known which native plants are best used for pollinator plantings in gardens. (Hallmann et al., 2017; Sanchez-Bayo Candidate plants must be attractive to pollinators, but they also must have and Wyckhuys, 2019). attributes that gardeners find appealing. To identify native plants that are Because of their ecological bene- attractive to gardeners, we disseminated two surveys. The first asked gardeners to fits, native plants are a market niche use a 5-point Likert scale to rate how likely they would be to garden with 23 that nursery plant growers and ven- fl fi owering plants native to the Paci c Northwest United States. The second dors could target. Currently, native survey asked gardeners to use a 5-point Likert scale to rate how likely they would plant gardening is an area of interest be to garden with a subset of 11 of these 23 native plants before and after receiving information about each flower’s attractiveness to bees (Anthophila). and growth in the landscaping and Using the first survey, we found a high level of acceptance of native plants by nursery industries. The American So- home gardeners (6 of 23 flowers had a mean “likelihood of planting” score ciety of Landscape Architects ranked of $ 4). Additionally, gardeners stated their likelihood of planting these native native plants and drought-tolerant species increased significantly after receiving information about the bees plants as the two most popular land- associated with each plant. Across both surveys, gardeners who identified as scape and garden trends in 2018 “native plant gardeners” stated they would be significantly more likely to garden (Howard, 2018), and the use of na- with all native plant species. Both surveys included an opportunity to share open- fl tive plants by landscape architects and ended comments, which revealed that gardeners were most concerned with ower contractors in the southeastern Unit- aesthetics and the aggressiveness of growth. Gardeners felt most positively about flower aesthetics and beneficial ecological traits. Many gardeners also commented ed States has increased with clientele that they needed more information or were unfamiliar with the plants. This interest (Brzuszek and Harkess, study shows that native plants can have high baseline appeal to home gardeners. 2009). Despite increasing public in- Specifically, we identified five native plant species that northwestern U.S. terest in incorporating native plants in nurseries might consider growing and marketing as pollinator plants because of gardens and other ornamental land- their high level of attractiveness to bees and home gardeners: globe gilia (Gilia scapes, these species are not widely capitata), california poppy (Eschscholzia californica), douglas aster Symphyotrichum subspicatum Eriophyllum lanatum planted by home gardeners and are ( ), oregon sunshine ( ), and not widely available at retail nurseries common yarrow (Achillea millefolium). (Avolio et al., 2018; Loram et al., 2008). It is unclear how public inter- est is balanced with the barriers to nterest in ecological gardening has that gardens composed of native plants adoption that exist at the consumer level and the production level. increased in recent years, largely are associated with higher native insect On the consumer level, a lack of driven by concerns over environ- b-diversity (the change in species di- I availability contributes heavily to the mental issues, including water use and versity between sites in a community) low rate of adoption of native garden the potential invasiveness of ornamen- than those planted with exotic species tal garden species (Kauth and P erez, plants (Wilde et al., 2015). Native plants (Burghardt and Tallamy, 2015). Other of limited diversity are available for sale 2011). A common theme of ecologi- studies have shown that exotic plant cal gardening efforts is the inclusion at nurseries, and 74% of the vascular species support lesser insect species plant species native to the United States of native plants in landscape design abundance than native plants (Bur- (Uren et al., 2015). Research suggests are not sold commercially (White et al., ghardt et al., 2010). Further work by 2018). Additionally, gardeners tend to Narango et al. (2017) suggested that Received for publication 7 Dec. 2020. Accepted for lack knowledge about native plants publication 16 Apr. 2021. urban plantings should include native (Kauth and P erez, 2011), thereby com- Published online 2 June 2021. plant species to provide improved hab- pounding the issue of availability be- 1Department of Horticulture, Oregon State Univer- itats and sustain native fauna. cause gardeners who are not aware of sity, 4017 Agriculture and Life Sciences Building, When considering bees (Antho- native plants will not request them to be Corvallis, OR 97331 phila) in particular, many species are stocked for sale. Finally, consumers tend A.G.A. is the corresponding author. E-mail: generalists and can be found broadly to prefer cultivated plants that have [email protected]. foraging across an array of native, non- been bred for aesthetic appeal, or gar- This is an open access article distributed under the CC BY-NC-ND license (https://creativecommons. native, and cultivated plant species (Bal- den standards that are familiar and tradi- org/licenses/by-nc-nd/4.0/). dock et al., 2015; Matteson and Lan- tionally used in ornamental landscapes https://doi.org/10.21273/HORTTECH04770-20 gellotto, 2011; Salisbury et al., 2015; (Yue et al., 2012).
458 August 2021 31(4) Multiple barriers to adoption also that are native to the northwestern tolerance, pollinator appeal, and abili- exist at the production level. Nurseries United States. The objective of our ty to grow in full sun. These filters re- will not increase their selection of native first survey was to quantify the base- sulted in a list of 63 potential native plants for sale unless a market exists for line appeal of 23 native northwestern plant species, and the final species these species, forming a feedback loop U.S plant species to gardeners. For were selected in cooperation with a with a lack of demand from consumers this first survey, we hypothesized that landscape designer who used her ex- (Wilde et al., 2015). From a produc- gardeners who identified themselves perience with these plants and her tion standpoint, there is limited infor- as a native plant gardener would find knowledge of past clients’ preferences mation regarding the performance of native plants more aesthetically ap- to examine an initially broad list and native plants across a realistic range of pealing and would be more willing to suggest 23 that had high potential for garden settings, thus complicating the include these native plants in their adoption by home gardeners. She also prioritization of plants to produce for gardens than other gardeners. noted that her past client choices were sale. Similarly, propagation methods The objective of the second sur- largely influenced positively by aes- for these plants are generally lacking, vey was to investigate whether access thetics and negatively by aggressive and smaller niche nurseries lack resour- to information regarding the ecologi- growth (S. Danler, personal communi- ces to create protocols or introduce cal benefits of specific native plants cation). We used the U.S. Department new plants into production (Norcini, would change gardeners’ opinions of of Agriculture (USDA) PLANTS on- 2007). In addition, the localized nature these plants. We hypothesized that all line database to confirmthenativesta- of native plants means that markets are gardeners would increase their positive tus of each species (USDA, 2016). regionalized. This can make it difficult impression of native plants by a similar These species are also being tested dur- for growers to scale-up production and magnitude, irrespective of demograph- ing a field study seeking to quantify can limit sales to smaller geographic ic group, after learning about the bees their attractiveness to pollinators, natu- regions. that are associated with various native ral enemies, and predators. If nursery growers are interested plant species. Again, we hypothesized SURVEY DEVELOPMENT. We as- in meeting the public’s demand for that gardeners who identify as native sessed the attitudes of gardeners to- native plants, then it would be useful plant gardeners would rank these plants ward 23 native plants using two online to know which plants show the most higher, overall, than non-native plant surveys created using a web-based sur- promise for retail sales and how gar- gardeners. vey tool (Qualtrics XM; Qualtrics, Pro- deners who identify as native plant vo,UT).Usingthefirst, or aesthetic gardeners (compared to the general Materials and methods preferences, survey (Supplemental Ma- gardening public) perceive native STUDY PLANTS. A total of 23 na- terial S1), we sought to ascertain the plants as garden plants. Therefore, we tive plant species (Table 1) were se- baseline appeal of native plants using conducted two online surveys to lected for this survey. These plants two metrics: perceived aesthetics and quantify the appeal of flowering plants were chosen based on their drought the gardeners’ willingness to include a
Table 1. List of 23 flowering plants native to the northwestern United States included on two surveys about native plant preferences of gardeners. The native range was determined using the PLANTS database (USDA, 2016). Common name Scientific name Annual/perennial Native range Smallflower lotus Acmispon parviflorus Annual Western United States Farewell-to-spring Clarkia amoena Annual Pacific Northwest United States Giant blue-eyed mary Collinsia grandiflora Annual Western United States Globe gilia Gilia capitata Annual Western United States Common sunflower Helianthus annuus Annual United States Miniature lupine Lupinus micranthus Annual Western United States Common madia Madia elegans Annual Western United States Baby blue eyes Nemophila menziesii Annual Western United States Varileaf phacelia Phacelia heterophylla Annual Western United States California poppy Eschscholzia californica Annual/perennial Western United States Common yarrow Achillea millefolium Perennial United States Pearly everlasting Anaphalis margaritacea Perennial United States Showy milkweed Asclepias speciosa Perennial United States Western columbine Aquilegia formosa Perennial Western United States Great camas Camassia leichtlinii Perennial Pacific Northwest United States Oregon sunshine Eriophyllum lanatum Perennial Western United States Wild strawberry Fragaria vesca Perennial United States Oregon iris Iris tenax Perennial Pacific Northwest United States Cream stonecrop Sedum oregonense Perennial Pacific Northwest United States Rose checkermallow Sidalcea asprella ssp. virgata Perennial Pacific Northwest United States Idaho blue-eyed grass Sisyrinchium idahoense Perennial Pacific Northwest United States Canada goldenrod Solidago canadensis Perennial United States Douglas aster Symphyotrichum subspicatum Perennial Western United States
August 2021 31(4) 459 particular plant in their own gardens. 23 native plants in individual experi- the “likelihood of planting” questions Institutional Review Board approval mental plots in the field, sampled the largely mirrored the responses to the was obtained for these surveys in July pollinator communities visiting them “attractiveness” questions. Therefore, 2018. using an insect vacuum, and observed we only reported the Likert results for We displayed photographs of each the abundance of pollinator visitors via “likelihood of planting” questions here native plant in bloom. These photo- timed counts. The 11 plants were com- because this question is more pertinent graphs were either obtained by us in mon yarrow (Achillea millefolium), to indicating potential adoption. We the field or licensed by Creative Com- pearly everlasting (Anaphalis margari- report the qualitative comments associ- mons (Mountain View, CA) and avail- tacea), douglas aster (Symphyotrichum ated with both sets of questions. able on the Internet. We included a subspicatum), farewell-to-spring (Clark- SURVEY ONE:AESTHETIC PREFER- close-up image of a representative flow- ia amoena), oregon sunshine (Eriophyl- ENCES. To determine whether there er and a second, more zoomed-out lum lanatum), california poppy was a difference in “likelihood of photograph of the flower with foliage if (Eschscholzia californica), globe gilia planting” scores among native plant the first photograph did not capture it. (Gilia capitata), common madia (Mad- species, we performed an analysis of TheseimageswerefollowedbytheLik- ia elegans), varileaf phacelia (Phacelia variance (ANOVA). Although Likert ert scale questions “on a scale of 1–5, heterophylla), rose checkermallow (Si- data are ordinal, not continuous, how attractive do you find this flower?” dalcea asprella ssp. virgata), and canada many statisticians believe it is accept- and “on a scale of 1–5, how likely goldenrod (Solidago canadensis). Sec- able to use parametric tests if the scale would you be to plant this flower in ond, after gardeners ranked each plant is 5 points or more points (Harpe, your home garden?”,where1wasthe according to their attractiveness and 2015) and the sample size is larger lowest score and 5 was the highest willingness to plant, they were shown than 15 (Mircioiu and Atkinson, score. Each of these questions also in- information that we had gleaned from 2 2017). Our sample size was sufficient- cluded a text box to allow respondents years of data collection (A.G. Anderson ly large and consisted of 5 points; to make open-ended comments about and G.A. Langellotto, unpublished therefore, we performed a parametric each species. data) as well as a photograph of a polli- type II partial sum of squares ANOVA The survey also included demo- nator on that flower. For example, using the “Anova” function in the graphic questions related to how long “Phacelia heterophylla attracted the “car” package (Fox and Weisberg, respondents have been gardening and highest abundance and species richness 2019). We did not perform post hoc whether they considered themselves of bumblebee species in 2018” (Supple- comparisons among the 23 native to be native plant gardeners. We also mental Material S2). After learning plant species during this study because asked respondents to identify barriers about the bees that were associated with adjusting the probability values for that would prevent them and other each native plant species, gardeners 253 pairwise comparisons would re- gardeners from adopting native plants were queried about whether they sult in extremely low statistical power. in garden settings (see Supplemental viewed the plant species more favorably, Instead, we used graphical displays to Material S1 for all questions). less favorably, or the same. Finally, gar- visually assess which plants were most The questions in the second, or denerswereaskedtoonceagaintorate attractive to gardeners using the ecological benefits, survey (Supple- how attractive they found the flowering “ggplot2” package (Wickham, 2016). mental Material S2) largely mirrored native plants and how likely they would The first graphical display was a bar the questions from the aesthetic pref- be to use the species in their gardens us- graph of the mean aesthetics Likert erences survey. Specifically, the demo- ing the 5-point Likert scale. scores for each plant species (Fig. 1). graphic questions were the same SURVEY DISTRIBUTION. Both sur- Our second graphical display was a di- (except for the addition of a question veys were distributed via e-mail list- verging stacked bar chart of the Likert asking about the gardeners’ loca- serves (Oregon Master Gardener) and scores, which made it easier to visual- tions). As in the aesthetic preferences via social media on Facebook (Menlo ize not only mean responses but also survey, we asked gardeners to rate Park, CA), Twitter (San Francisco, the distribution of the positive and each native plant species for attractive- CA), and Instagram (Menlo Park, negative ratings of native plant species ness and their willingness to include CA) targeting Oregon gardeners. (Fig. 2). this plant in their gardens using a 5- These methods were used to reach as Additionally, we were interested in point Likert scale (1 = lowest score; many gardeners as possible, particu- differences in the responses of native 5 = highest score). However, there larly those who reside in the north- plant gardeners and non-native plant were a few key differences between western United States. The aesthetic gardeners. We used a Welch’s t test the two surveys. First, instead of ask- preferences survey was open to re- (0.05 alpha level) to examine differ- ing gardeners about their impressions sponses between 10 July 2018 and 7 ences in mean responses between these of the full set of 23 native plants, we Jan. 2019. The ecological benefits two demographic groups. All t tests in instead focused on 11 plants that survey was open to responses between this study were performed using the showed promise as highly attractive 20 May and 25 Sept. 2019. “t.test” function in the R Base Package. pollinator plants based on a broader ANALYTICAL METHODS. We per- SURVEY TWO:ECOLOGICAL BEN- field experiment we were conducting formed all statistical analysis using EFITS. We used a paired t test to to assess these plants’ attractiveness to RStudio (RStudio, Boston, MA) and evaluate whether the “likelihood of insect visitors (A.G. Anderson and the statistical software R (version 3.5.1; planting” scores were significantly dif- G.A. Langellotto, unpublished data). R Core Team, Vienna, Austria). Across ferent for all gardeners before and af- During this experiment, we planted all both surveys, the Likert responses to ter the information related to the bees
460 August 2021 31(4) associated with each native plant was shared with participants. This allowed us to determine whether gardeners change their opinions after learning the ecological benefits of plant spe- cies. Again, with a large sample size, t tests are robust for non-normal Likert data. We again used Welch’s t tests (0.05 alpha level) to determine if there was a statistically significant dif- ference in the mean “likelihood of planting” responses of native plant gardeners and non-native plant gar- deners before and after learning about the benefits to pollinators that these plants provide. Finally, we calculated the differ- ences between the before and after Likert scores for “likelihood of plan- Fig. 1. Mean score (5-point Likert scale) of the likelihood respondents would ting.” We used Welch’s t tests (0.05 garden with each of 23 plant species native to the Pacific Northwest United alpha level) to compare the mean States, as ranked by respondents during an online survey (aesthetic preferences) score changes of native plant garden- investigating the appeal of these species to home gardeners. A Likert score of 1 ers relative to those of nonnative plant represents the least attractive and a Likert score of 5 represents the most attractive. gardeners to detect whether the mag-
Fig. 2. Percentage of responses to the question “how likely would you be to plant this flower?” using five Likert scores of an online survey (aesthetic preferences) investigating the appeal of plant species native to the PacificNorthwestUnited States to home gardeners. A Likert score of 1 represents the least attractive and a Likert score of 5 represents the most attractive.
August 2021 31(4) 461 Table 2. Definitions of the codes used to categorize the open-ended comments addition, each text response was classi- from two online surveys investigating the native plant preferences of home garden- fied as positive, negative, or neutral in ers. The comments refer to the aesthetic appeal of the native plants to the respond- reference to a theme. For example, ents and the likelihood that the respondents would use these native plants in their “ ’ the response would not plant this, home gardens. Codes help us understand patterns in gardeners plant choices. deer eat it” would be categorized as Code Definition negative and pests. The response “I fl ” Negative comments like purple owers would be catego- Aesthetics Negative impression of aesthetics (e.g., color, size, foliage rized as positive and aesthetics. shape, etc.) After scoring the qualitative re- Aggressiveness/ Concerns regarding aggressive growth or self-seeding sponses according to theme and con- weediness notation (e.g., positive or negative), Yard habitat Gardener yard habitat is not appropriate for the plant we counted the number of comments Pest concerns Concern that a species will attract pests [i.e., pest insects, with each code to determine the most deer (Cervidae), voles (Microtus sp.)] positive and negative factors influenc- Establishment/ Pertaining to plant establishment/growth (i.e., slow, difficult ing respondents’ impressions of native growth to establish) plant species. Responses with multiple Availability Unlikely to use because of low availability themes were assigned to each code No space Comment reflects a lack of space in which to grow this plant that was appropriate. Allergies Comment reflects concerns about allergies Phenology Grows or blooms at a disagreeable time Results Not native Gardeners claimed that the plant is not native to their region DEMOGRAPHICS. A total of 482 Poisonous Concern about plant toxicity [e.g., showy milkweed people fully or partially completed the (Asclepias speciosa)] aesthetic preferences survey. Of these, Cross-pollination Concern regarding plants cross-pollinating with other plants 64% considered themselves native in/near garden plant gardeners, 33% did not, and 3% General General negative impression did answer the question related to Cultural needs Comments relating to concerns over cultural care practices needed whether they considered themselves to Smell Disagreeable smell be a native plant gardener (Table 3). Positive comments A total of 334 people completed Already have Gardeners indicated they already garden with this plant the ecological benefits survey. Of these, Aesthetics Positive impression of aesthetics (e.g., color, size, foliage 44% considered themselves native plant shape, etc.) gardeners, 22% did not, and 34% did Ecology Positive impression of the plant’s ecological benefits (i.e., not answer the native plant gardener attracting wildlife, drought tolerance) question. Only 11% of respondents Edible/medicinal Positive views of the plant’s use for food or medicinal stated that they had completed the purposes previous aesthetic preferences survey Ease of care Easy to care for (i.e., easy propagation, low water needs, (Table 3). hardiness) SURVEY ONE:AESTHETIC PREFE- Pest resistance Positive impression of the plant’s resistance to pests (insect RENCES. The first survey asked garden- pests, deer) ers to rankthe attractiveness and thelike- Phenology Grows or blooms at an agreeable time lihood they would plant each of the 23 General General positive impression native plant species. Gardener rankings Smell Agreeable smell of the likelihood they would use each na- Neutral comments tive plant differed significantly (at the Need more Gardeners stated the need for more information to decide 0.05 level) among flower species (P < information 0.001). Although there was variance in I would if … Gardeners stated they would use this plant if certain gardeners’“likelihood of planting” conditions were met (Fig. 1), visual inspection of the data Unfamiliar with Gardeners are not familiar with this plant showed that the difference between plant most native plant species was fairly small. General Neutral general impression The overall difference between the species with the highest score, western columbine [Aquilegia formosa (mean nitude of change was influenced by qualitative data, we coded themes Likert score = 4.61)] and the lowest the short messages about pollina- found in open responses for each sur- score, varileaf phacelia (mean Likert tor–plant associations. vey. Themes included the range of re- score = 2.73) is moderate, differing by SURVEY COMMENTS. Both sur- corded responses and allowed us to 41% on a 5-point Likert scale. Only veys generated qualitative results via aggregate responses in a manner that four native plant species had “like- open-ended response boxes that al- enabled us to assess potential patterns. lihood” scores less than 3.5 [rose checker- lowed respondents to reply with free- These themes included aesthetics, mallow, canada goldenrod, smallflower form comments related to each native ease of care, pest concerns, ecology, lotus (Acmispon parviflorus), and varileaf plant species. To analyze these and aggressiveness (Table 2). In phacelia]. Six of the flowers had a mean
462 August 2021 31(4) Table 3. Distribution of respondents who participated in two online surveys 238 were coded neutral (Table 4). Re- [aesthetic preferences (N = 482) and ecological benefits (N = 334)] investigating garding the negative comments, the the native plant preferences of home gardeners who identify as native plant gar- two most common responses were deners. The second column displays the percentage of gardeners who participated negative impressions of the flowers’ in both the first survey and the second survey. aesthetics (N = 184) and concerns Respondents who participated about the aggressiveness of plant Respondents who were in the aesthetics preferences growth (N = 121). The theme of the Responses native plant gardeners (%) survey (%) most comments was “already have” Aesthetic preferences survey (meaning the gardener indicated that Yes 63.7 — they currently have this plant species in No 33.3 — their garden currently or have gardened No response 3.0 — with it in the past) (N = 627), whereas fi Ecological bene ts survey positive impressions of aesthetics (N = Yes 44.0 11.0 107) and beneficial ecological traits No 22.2 51.2 (N = 52) were the most common rea- No response 33.8 37.8 sons given for positive impressions. In the neutral category, needing more in- formation about a plant was the most “likelihood” score of $4.0 using the groupings of plants that were more common response (N = 172), followed 5-point Likert scale. These were western likely to be planted or that gardeners by “I would if …” (N = 38), which columbine (4.61), oregon iris [Iris tenax were unlikely to plant (Likert scores represented gardeners who liked the (4.45)], idaho blue-eyed grass [Sisy- of 1 or 2), other than a noticeable in- plants but expressed reasons why they rinchium idahoense (4.33)], great camas crease in lower scores for the two would not use them. [Camassia leichtlinii (4.25)], globe gilia plants with the lowest scores, small- Farewell-to-spring, showy milk- (4.56), and cream stonecrop [Sedum flower lotus and varileaf phacelia (Fig. weed, and california poppy were the oregonense (4.15)]. Another six species 2). Visually, canada goldenrod ap- species that received the most overall had scores between 3.8 and 4.0: baby peared to be polarizing, with fewer comments. Showy milkweed, california blue eyes [Nemophila menziesii (3.99)], scores of 2, 3, and 4 than similarly poppy, and western columbine had the showy milkweed [Asclepias speciosa ranking plants, and more scores of 1 most positive comments, whereas can- (3.97)], douglas aster (3.92), common and 5. Of those who participated in ada goldenrod, farewell-to-spring, and yarrow (3.87), common sunflower [Hel- our survey, native plant gardeners (N showymilkweedhadthemostnega- ianthus annus (3.83)], and california =307)weresignificantly more likely tive comments (Fig. 3A). poppy (3.81). There were no obvious to be willing to plant the species than Because so many gardeners stated groupings of plants that were “more non-native plant gardeners (N = 161) that they had currently or previously likely” or “less likely” to be planted (P < 0.001). gardened with any of these native based on visual inspection of the mean CODED COMMENTS. There were plants, we counted the number of Likert plots (Fig. 1). 1643 individual comments in the aes- “already have” comments each re- Although most plants were rela- thetic preferences survey across both ceived (Table 5). Common yarrow, tively appealing to gardeners, visual the “attractiveness” and “likelihood” western columbine, great camas, cali- inspection of the diverging stacked questions. Of these, 818 were coded as fornia poppy, and oregon iris all had bar chart revealed no obvious positive, 587 were coded negative, and more than 50 gardeners’ comments,
Table 4. Number of respondent comments for each topic from a survey of 23 flowering plants native to the northwestern United States with the potential for use in home gardens. These comments were open-ended and referred to native plants’ aesthetic appeal and likelihood that respondents would plant these species in their own garden (N = 482). Negative Responses (no.) Positive Responses (no.) Neutral Responses (no.) Aesthetics 184 Already have 627 Need more information 172 Aggressiveness 121 Aesthetics 107 I would if … 38 Yard habitat issue 57 Ecology 52 Unfamiliar with plant 26 Establishment/growth 48 Ease of care 13 General 1 Availability 44 Edible/medicinal 8 Phenology 1 Pest concerns 30 Phenology 6 No space 27 Pest resistance 3 Allergies 23 General 2 Phenology 23 Not native 10 Poisonous 7 Cross pollination 4 General 2 Cultural needs 7
August 2021 31(4) 463 Fig. 3. (A) Number of negative, positive, and neutral open-ended comments during an online survey (aesthetic preferences) investigating the appeal of 23 plant species native to the Pacific Northwest United States to home gardeners. (B) Number of negative, positive, and neutral gardener open-ended comments during a second online survey (ecological benefits) investigating the role of ecology in the appeal of 11 plant species native to the Pacific Northwest United States to home gardeners.
464 August 2021 31(4) Table 5. Number of comments from unprompted, indicating that they al- plant gardeners (N = 147) had signifi- respondents indicating that they had ready used these species. cantly higher likelihood of planting currently or previously gardened with SURVEY TWO:ECOLOGICAL BEN- Likert scores for all study plants com- the native plant species listed on the fl EFITS. After receiving information pared with those who did not identify survey of 23 owering plants native as native plant gardeners (N = 74). to the northwestern United States about the ecological attributes of study with the potential for use in home plants, gardeners stated that they would This was the case for scores before gar- gardens (N = 482). be significantlymorelikelytoplant deners received the information about them overall [P < 0.001 (Fig. 4)]. Ac- bees associated with each plant (P < Plant species Responses (no.) cess to information about the bees asso- 0.001) as well as after receiving this in- Common yarrow 59 ciated with each native plant increased formation (P < 0.001). Western columbine 57 the overall number of native plants Compared with non-native plant Great camas 52 with “likelihood of planting” Likert gardeners, self-identified native plant California poppy 50 scores $4 from three plants to eight gardeners did not exhibit a significantly Oregon iris 50 plants (Fig. 4). Varileaf phacelia had different change in how likely they Showy milkweed 43 the largest jump, with an increase in would be to plant these species after Idaho blue-eyed grass 40 mean scores of >1.0 using the Likert learning about their bee visitors (P = Douglas aster 38 scale. 0.085). Cream stonecrop 37 Although the absolute Likert SURVEY COMMENTS. There were Rose checkermallow 33 scores increased across all plants in- 485 individual comments in the ecolo- Wild strawberry 32 cluded in this second survey (Fig. 4), it gical benefits survey across the “attrac- Oregon sunshine 26 is interesting to note that visual inspec- tiveness” and “likelihood of planting” Canada goldenrod 24 tion of the relative positioning of na- questions. Of these, 254 were coded as Common sunflower 20 tive plants to each other revealed only positive, 175 were coded as negative, Farewell-to-spring 16 minor reshuffling of each plant’srank. and56werecodedasneutral(Table6). Pearly everlasting 14 Plants that were of a lower rank before As in the first survey, the two most com- Globe gilia 9 the ecological information was shared mon negative responses were related to Miniature lupine 8 maintained a relatively low rank after anativeplant’saesthetics(N=57)and Varileaf phacelia 8 gardeners learned more about the eco- concerns about the aggressiveness of Common madia 5 logical attributes of the plant. plant growth (N = 34). Again, the code Giant blue-eyed mary 3 COMPARISON OF NATIVE PLANT “already have” (N = 195) was the most Baby blue eyes 3 GARDENERS TO OTHER GARDENERS. common positive theme, whereas posi- fl Small ower lotus 0 Respondents who identified as native tive impressions of aesthetics (N = 27)
Fig. 4. Mean score (5-point Likert scale) of the likelihood that respondents would garden with each of 11 plant species native to the Pacific Northwest United States before being provided with information about the pollinators that visit these plants displayed next to the mean “likelihood of planting” score by native plant species after learning about these plants’ pollinator visitors.
August 2021 31(4) 465 Table 6. Number of respondent comments for each topic from a survey of 11 flowering plants native to the northwestern United States known to be attractive to pollinators. These comments were open-ended and referred to native plants’ aes- thetic appeal and the likelihood that respondents would plant these species in their own garden (N = 334). Negative Responses (no.) Positive Responses (no.) Neutral Responses (no.) Aesthetics 57 Already have 195 Need more information 41 Aggressiveness 34 Aesthetics 27 I would if … 13 Yard habitat issue 24 Ecology 20 Unfamiliar with plant 2 Availability 19 Ease of care 4 Establishment/growth 11 Phenology 4 No space 10 Edible/medicinal 3 Allergies 7 General 1 Phenology 5 Not native 5 Pest concerns 2 Smell 1 and beneficial ecological traits (N = 20) promise for the retail nursery market included in the second survey after be- were the most common reasons given were western columbine, oregon iris, ing exposed to brief messages about for positive impressions. In the neutral idaho blue-eyed-grass, great camas, the bees associated with each plant. category, needing more information globe gilia, and cream stonecrop. The increases we observed in about a native plant was the most com- These all had scores of $4.0 using the these scores suggested that a signifi- mon response (N = 41), followed by “I 5-point Likert scale. If we included cant subset of gardeners consider eco- would if …” (N = 13) (representing results from the second survey, then logical traits a component of a gardeners who liked the plants but ex- california poppy, oregon sunshine, co- flower’s appeal, and that a minimal pressed reasons they would not use mmon yarrow, common madia, fare- amount of education can significantly them). well-to-spring, douglas aster, and rose sway gardeners’ opinions of native Farewell-to-spring again received checkermallow also scored $4.0 for plants. Based on these results, nurser- the most comments overall, followed “likelihood of planting” after garden- ies and garden stores may be able to by common yarrow and california ers knew the attractiveness of these increase the palette of native plants poppy. Common yarrow and califor- plants to bees. that gardeners find attractive and suit- niapoppyreceivedthemostpositive It is notable that in these surveys, able for planting by sharing brief mes- comments, whereas canada goldenrod gardeners indicated acceptance for sages about the ecological benefits and rose checkermallow received the flowers that are not known for, and these plants provide. Previous research most negative comments (Fig. 3B). have not been bred for, their aesthetic has shown that labels result in con- beauty. Previous research has found sumers paying a premium for native Discussion that half of the surveyed gardeners said plants (Yue et al., 2011, 2012); there- Ultimately, we were interested in that attracting pollinators was a reason fore, labeling might benefitgrowers identifying native plants that garden- why they use pollinator-friendly plants by increasing the quantity of native ers would be willing to incorporate in (Campbell et al., 2017). This suggests plants sold and their individual prices. ornamental garden settings on their that focusing marketing on both the Our findings that the stated likeli- properties. This information can be ecological and aesthetic values of na- hood that gardeners will use native shared with gardeners to suggest spe- tive plants might be advantageous to plants can be increased by small cific native plants that both support gain attention from the subset of gar- amounts of education are encourag- pollinators and work well in residen- deners who find them pretty, as well as ing for efforts to increase ecological tial landscapes. These recommenda- those that are interested in promoting function in urban and suburban land- tions can also help growers and sellers pollinator habitats. scapes and highlight the importance of native plants to target species that The ecological benefits survey of outreach and education to the suc- fitgardeners’ aesthetic and ecological suggested that sharing a small amount cess of these efforts. Even if people are preferences. of information about the benefits pro- willing to take ecology into consider- Our results suggest that native vided by bee-friendly plants can signif- ation, they will not plant ecologically plants can have high baseline accep- icantly increase how likely gardeners friendly flowers if they are unaware of tance by home gardeners, thus repre- would be to use these plants. In this the benefits of these plants. In addition, senting a potentially profitable market second survey, only 3 of the 11 plants gardeners’ perceptions of “pollinator- for nursery growers. However, not all included in the survey initially had friendly” flower traits vary and are fre- native plants are likely to be equally mean “likelihood of planting” Likert quently incorrect (Khachatryan and embraced by the consumer market. scores more than 4.0. After learning Rihn, 2018). Although some gardeners We found significant differences in about the bees associated with each select traits like pollen production and how likely gardeners would be to use native plant species, eight plants had nectar production (generally good for our 23 study plants. Based on the re- mean scores more than 4.0. Notably, pollinators), others gravitate toward sults of the aesthetic preferences sur- gardeners stated that they would be bright flowers and brightly colored fo- vey, the plants that showed the most more likely to use all 11 native plants liage, which are traits that are often not
466 August 2021 31(4) beneficial to pollinators (Khachatryan with allergenic ragweeds (Ambrosia Many respondents to the aesthetic and Rihn, 2018). Existing ideas and sp.). These comments further reinforce preferences survey commented that knowledge frequently bias gardeners’ the need for outreach and education they already used some of these native impressions of what a pollinator-friend- on behalf of native plants. plants in their home gardens. Of the ly plant is (Campbell et al., 2017; Wol- Moving forward, these results candidate native plants we recommend, laeger et al., 2015). can inform the sale of native plant ma- common yarrow had 59 “already have” Not surprisingly, the respondents terials on the retail market. There is comments, and california poppy had who identified as native plant garden- growing interest in using native plants 50 “already have” comments. This may ers were significantly more likely to in ornamental landscapes, but most suggest that these two species already find native plants attractive, both be- gardens are comprised of exotic have higher market availability. How- fore and after learning about the polli- plants. For example, Smith et al. ever, douglas aster (N = 38), oregon nator attractiveness of these plants. It (2006) found that garden flora in sunshine (N = 26), and globe gilia (N = stands to reason that gardeners who Sheffield, England, comprised 70% 9) had substantially fewer of these com- are already aware of these species and exotic plants. Contributing to this is a ments. This further reinforces the idea enjoy using them would be more ap- lack of availability because many larger that these plants have an opportunity preciative of their aesthetics. “box” garden stores have few native for nurseries to consider propagating Our findings are encouraging, es- species available for sale (Avolio et al., and marketing. pecially because there were various 2018; Wilde et al., 2015). Additional- reasons why gardeners may or may ly, native plant species may have traits Limitations not include a plant in their garden. Al- deemed to be undesirable in gardens, It should be noted that potential though we did not share the “attr- as reflected in our survey comments. biases existed during these surveys. activeness” scores from our survey Native plants that are not tradi- Self-selection was present because of (because they tracked the same patterns tionally aesthetically pleasing may still our recruitment methods (Bethlehem, observed for the “likelihood of plan- have a place in the modern ornamental 2010). Both surveys were distributed ting” scores), the mean “attractiveness” landscape. Cubino et al. (2020) sur- via e-mail listservs and social media Likert scores were slightly, but consis- veyed homeowner vegetation and channels to reach as many gardeners as tently, higher than the mean “like- management choices for 145 yards possible at low cost. Therefore, we did lihood” scores. This score gap is not across the United States and found that not randomly select participants. Per- surprising because gardeners consider landscape priorities were largely sepa- haps gardeners who participated were more factors than just aesthetics when rated into four categories: natural, neat, largely from certain demographics that deciding what to plant. showy, and low-cost. This revealed that may influence their views of native The coded comments from the gardeners do have different priorities, plants. Previous research has found that aesthetic preferences survey provide and our gardener comments tracked the adoption of ecologically friendly insight regarding some of the factors these results. Appropriate native plants practices is influenced by social norms and concerns that may drive garden- may be uniquely able to cross these cat- (Bamberg, 2003). Age, education, and ers’ native plant choices and the gap egories with broad appeal. They can be location are associated with the adop- between how attractive gardeners find marketed for use in native or “natural” tion of environmentally friendly practi- plants and their willingness to plant landscaping, can be drought-tolerant ces, but not gender or race (Dietz them. These survey comments re- and need few soil amendments (“low- et al., 1998), and it may be that similar vealed that, second only to aesthetics, cost” gardens), and many species are dynamics influenced this study. the aggressiveness or weediness of na- attractive (for use in “showy” gardens). Additionally, only gardeners who tive plants was a major concern to gar- To increase the promise of the na- had access to the Internet and were able deners. Although these comments tive plant market, nurseries should to read English could participate in our appeared less often, concerns about provide increased availability of plants surveys. In general, those with Internet successful plant establishment, not that fit the desired criteria, such as aes- access often answer questions differently having an appropriate yard, and po- thetics, growth habit, and size. To do than those without (Bethlehem, 2010); tential lack of availability were men- so effectively, native plants marketed therefore, these results were biased to tioned by gardeners. for sale by nurseries and garden stores the responses of a population with web Although the survey respondents may need to clear a baseline appeal access. Similarly, English-speaking gar- were presumably highly educated threshold to home gardeners and deners encompass demographics with about garden topics (many were Mas- should be coupled with messages that certain viewpoints and values that may ter Gardeners, N = 523 across both extol the nonaesthetic benefits of par- differ from those of some non-English- surveys), a substantial number of com- ticular plants. Based on the results of speaking demographics. ments referred to needing more infor- our surveys and the results from our The survey questions were not mation or being unfamiliar with the companion field study (A.G. Anderson presented in a new random order to plants. Additionally, gardeners had and G.A. Langellotto, unpublished each respondent, which might have misconceptions about some of these data), we suggest that globe gilia, cali- introduced bias or survey fatigue as plants, which may have reduced their fornia poppy, douglas aster, oregon confounding variables when interpret- scores. For example, canada goldenrod sunshine, and common yarrow are ing the results. People may have be- received multiple comments about it good candidates for plants that have come tired of responding by the end causing allergies, and it is likely that high aesthetic appeal and can also be of the survey; therefore, they might gardeners were confusing this species marketed as bee-friendly plants. have ranked the plants at the end
August 2021 31(4) 467 consistently higher or lower. Howev- Baldock, K.C., M.A. Goddard, D.M. Hicks, Hallmann, C.A., M. Sorg, E. Jongejans, er, a visual display of the mean Likert W.E. Kunin, N. Mitschunas, L.M. Osga- H. Siepel, N. Hofland, H. Schwan, W. scores displayed in the order of the sur- thorpe, S.G. Potts, K.M. Robertson, A.V. Senmans, A. Muller,€ H. Sumser, T. € vey did not indicate a detectable pat- Scott, G.N. Stone, and I.P. Vaughan. 2015. Horren, D. Goulson, and H. de Kroon. ’ tern (data not shown). To further Where is the UK s pollinator biodiversity? 2017. More than 75 percent decline over fl 27 years in total flying insect biomass in verify this, we performed a t test for The importance of urban areas for ower- fi visiting insects. Proc. Biol. Sci. 282:2014 protected areas. PLoS One 12:e0185809, the mean scores of the rst half and 2849, doi: 10.1098/rspb.2014.2849. doi: 10.1371/journal.pone.0185809. second half of the survey and found no significantdifferencebetweenthetwo Bamberg, S. 2003. How does environmen- Harpe, S.E. 2015. How to analyze Likert tal concern influence specific environmental- and other rating scale data. Curr. Pharm. (data not shown). However, placing – the demographic section at the end of ly related behaviors? A new answer to an old Teach. Learn. 7:836 850, doi: 10.1016/ – j.cptl.2015.08.001. the second survey (33.8% nonresponse question. J. Environ. Psychol. 23:21 32, rate for the question asking if they are doi: 10.1016/S0272-4944(02)00078-6. Hostetler, N.E. and M.E. McIntyre. native plant gardeners) instead of the Bethlehem, J. 2010. Selection bias in web 2001. Effects of urban land use on polli- beginning, as in our first survey (3.0% surveys. Intl. Stat. Rev. 78:161–188, doi: nator (Hymenoptera: Apoidea) communi- 10.1111/j.1751-5823.2010.00112.x. ties in a desert metropolis. Basic Appl. nonresponse rate for the same ques- – tion) was likely the cause of the differ- Ecol. 2:209 218, doi: 10.1078/1439- Brzuszek, R.F. and R.L. Harkess. 2009. 1791-00051. ences in responses to these questions. Green industry survey of native plant We could not calculate a response marketing in the southeastern United Howard, C. 2018. ASLA survey: Demand rate because it was impossible to know States. HortTechnology 19:168–172, doi: high for sustainable residential landscapes. 12 how many gardeners received the sur- 10.21273/HORTSCI.19.1.168. Oct. 2020.
468 August 2021 31(4) exotic species? J. Appl. Ecol. 52:1156–1164, U.S. Department of Agriculture. 2016. Appl. Ecol. 12:332–341, doi: 10.1016/j. doi: 10.1111/1365-2664.12499. The PLANTS database. 10 Dec. 2016. baae.2010.11.008.
August 2021 31(4) 469 Supplemental Material S1 d. Upkeep 22. Sisyrinchium idahoense. Common This survey was designed to ascertain the e. Other name: Idaho blue-eyed-grass baseline aesthetic appeal of 23 different native 23. Solidago canadensis. Common Willamette Valley, Oregon, wildflowers to Plant Aesthetics and name: Canada goldenrod gardeners, as well as their willingness to in- Willingness to Plant Questions clude these species in their own home Gardeners were shown photos of the Supplemental Material S2 gardens. following plants and asked to rank them This survey was designed to assess how Our study is investigating the relative at- on a scale of 1 (lowest) to 5 (highest), in malleable gardener opinions of native plants tractiveness of Willamette Valley, Oregon, terms of (a) each plant’s attractiveness and are after being educated about the benefits fl wild owers to pollinators and natural ene- (b) their willingness to include this plant they provide to pollinators. This survey con- mies. We hope that pollinator-friendly in their garden. Gardeners were also of- sists of a subset of the most attractive polli- plants will be used by home gardeners, thus fered the opportunity to provide additional nator plants from the 2017 and 2018 field providing habitat for these insects in urban comments about each plant, based upon seasons. and suburban landscapes. We are interested the provided photo. We are studying the relative attractiveness in your opinion on the ornamental value of of Willamette Valley, Oregon, native wild- these native plants that could be planted in fl 1. Clarkia amoena. Common name: owers to insect pollinators and natural ene- western Oregon gardens. mies. We are also interested in understanding Farewell-to-spring fl the potential value of these plants to home gar- Demographic Questions 2. Collinsia grandi ora. Common deners and landscapers. Thus, we would like name: Giant blue-eyed mary 1. Are you over 18 years old? to know your opinion on the ornamental beau- 3. Gilia capitata. Common name: a. Yes ty of these plants. We would also like to know Blue headed gilia whether or not the ecological benefits provid- b. No 4. Lupinus microcarpus. Common ed by various plants might influence opinions 2. Are you currently a Master Garden- name: Miniature lupine related to these same plants. er, or have you been a Master Gar- 5. Madia elegans. Common name: dener in the past? Demographic Questions Common madia a. Yes 6. Nemophila menziesii. Common 1. Are you over 18 years old? b. No a. Yes name: Baby blue eye 3. How long have you been a garden- b. No 7. Eschscholzia californica. Common er? Please answer to the nearest 2. What area best represents where name: California poppy year, such as “10 years” or “1 year.” you garden? 8. Helianthus annuus. Common 4. Would you consider yourself a a. Willamette Valley, Oregon name: Common sunflower ”native plant gardener”? b. Another region within Oregon 9. Phacelia heterophylla. Common a. Yes c. Outside of Oregon name: Varied-leaf phacelia b. No d. Outside of the United States 10. Acmispon parviflorus. Common 5. Do you currently have native plants 3. Did you participate in our previous name: Smallflower lotus in your garden? If so, what kinds? survey on native plant aesthetics? 11. Achillea millefolium. Common Select all that apply. a. Yes a. Trees name: Common yarrow b. Unsure b. Shrubs 12. Anaphalis margaritacea. Common c. No c. Herbaceous annuals name: Pearly everlasting 4. Are you currently a Master Garden- d. Herbaceous perennials 13. Asclepias speciosa. Common er, or have you been a Master Gar- e. Bulbs name: Showy milkweed dener in the past? f. Other 14. Aquilegia Formosa. Common a. Yes 6. What barriers might prevent YOU name: Western columbine b. No from including more native plants in 15. Symphyotrichum subspicatum, 5. How long have you been a garden- ’ their garden? Select all that apply. Common Name: Douglas aster er? Please answer to the nearest a. Cost 16. Camassia leichtlinii. Common name: year, such as “10 years” or “1 year.” b. Availability Great camas 6. Would you consider yourself a c. Aesthetics 17. Eriophyllum lanatum. Common “native plant gardener”? d. Upkeep name: Oregon sunshine a. Yes e. Other 18. Fragaria vesca. Common name: b. No 7. What barriers do you think might Woodland strawberry 7. Do you currently have native plants prevent OTHER GARDENERS 19. Iris tenax. Common name: Oregon in your garden? If so, what kinds? from including more native plants in iris Select all that apply. their garden? Select all that apply. 20. Sedum oregonense. Common a. Trees a. Cost name: cream stonecrop b. Shrubs b. Availability 21. Sidalcea asprella ssp. virgata.Com- c. Herbaceous annuals c. Aesthetics mon name: Rose checkermallow d. Herbaceous perennials
470 August 2021 31(4) e. Bulbs study and asked to rank them on aClarkia scale of amoena1. . Common name: Farewell-to-spring f. Other 1 (lowest) to 5 (highest), in terms of (a) 2. Gilia capitata. Common name: ’ 8. What barriers might prevent YOU each plant s attractiveness and (b) their will- Blue headed gilia ingness to include this plant in their garden. from including more native plants in 3. Madia elegans. Common name: Gardeners were also offered the opportunity their garden? Select all that apply. Miniature lupine a. Cost to provide additional comments about each plant, based upon the provided photo. 4. Eschscholzia californica. Common b. Availability For each plant, gardeners were then name: California poppy c. Aesthetics shown a photo of a bee visiting each 5. Phacelia heterophylla. Common d. Upkeep fl ower accompanied by a fact from our name: Varied-leaf phacelia e. Other first two years of field studies (e.g., 6. Achillea millefolium. Common 9. What barriers do you think might “Phacelia heterophylla was the most at- prevent OTHER GARDENERS tractive flower to bumble bees in both name: Common yarrow from including more native plants in the first and second years of our study.”). 7. Anaphalis margaritacea. Common their garden? Select all that apply. We then asked gardeners whether they name: Pearly everlasting a. Cost viewed this flower as more attractive, and 8. Symphyotrichum subspicatum. b. Availability again asked them to rank on a scale of 1 Common name: Douglas’ aster c. Aesthetics (lowest) to 5 (highest), in terms of (a) 9. Eriophyllum lanatum. Common ’ d. Upkeep each plant s attractiveness and (b) their name: Oregon sunshine e. Other willingness to include this plant in their 10. Sidalcea asprella ssp. Virgata. garden after learning of the ecological Common name: Rose checker- benefit to pollinators it provides. Plant Aesthetics and mallow Willingness to Plant Questions Opportunity to provide additional com- ments about each plant, based upon the 11. Solidago canadensis. Common Gardeners were shown photos of a sub- provided photo. name: Canada goldenrod set of ten plants from survey one that had
been attractive to pollinators during our field
August 2021 31(4) 471