Background

This study explored contemporary urban residents’ perspectives and views on, and attitudes towards trees and woodlands in the urban landscape.

A three-stage approach for the overall research included:

1. Four focus groups (2 in England and 1 each in Wales and Scotland) to support the design of a large-scale survey.

2. A large-scale online national survey covering England, Scotland and Wales of 6,000 respondents.

3. Four validation focus groups (2 in England and 1 each in Wales and Scotland) to explore patterns and trends from the survey results in more depth and detail. This dashboard presents results from the online survey. Survey participants were selected by city size, ethnicity, education and age to represent contemporary urban populations. The sample included 4,400 in England, 1,000 in Scotland and 600 in Wales with survey participants recruited via online panels.

The research questions for the survey included:

1. How are urban trees in different context perceived by contemporary urban residents throughout Great Britain?

2. Do sociodemographic factors such as age, gender race, class and socio-economic group affect the way people perceive urban trees?

A number of hypotheses were explored and focused on for example, difference in attitudes by demographics, difference in nature connection and appreciation of nature, differences in views about how urban tree information should be communicated, and willingness to get involved in urban tree management.

Statistical Methodology

Analysis was conducted in R (version 4.0.2, R Core Team 2020), with graphics produced using ggplot2 (Wickham, 2016) and plotly (Sievert, 2020) in R as follows:

Each question was analysed separately with responses converted into binary outcomes as described throughout. The ‘lme4’ package (Bates et al, 2015) was used to fit generalized mixed linear models. Binary responses were the outcome variable with age in years (continous), gender, employment, education, garden (yes/no), country of residence, ethnicity and dependents (yes/no) as fixed effect covariates. Where there were multiple sub-parts to a question (e.g. multiple choice or ratings for several statements), the sub-questions were added as a fixed effect covariate and participant ID as a random effect. The sub-question was also fit as an interaction with each fixed effect demographic factor and Type-II Anova tables used to test their significance. As the survey contained 18 questions to analyse a Bonferroni correction was applied to create a more stringent threshold for statistical significance (0.05/18 = 0.0028). Non-significant interactions were removed from the model (p > 0.0028) but all demographic variables retained as main effects.

The ‘emmeans’ package (Lenth, 2020) was used to extract the adjusted marginal means for any significantly associated fixed effects and interactions for each question. The cld function in the ‘multcomp’ package allowed pairwise comparisons to be made between all groups and the Tukey method used to correct for multiple comparisons (Hothorn et al, 2008).

Model fit for all regression models were assessed using the ‘DHARMa’ package (Hartig, 2020).

References

H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2016.

Douglas Bates, Martin Maechler, Ben Bolker, Steve Walker (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01.

Russell Lenth (2020). emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.5.1. https://CRAN.R-project.org/package=emmeans

Torsten Hothorn, Frank Bretz and Peter Westfall (2008). Simultaneous Inference in General Parametric Models. Biometrical Journal 50(3), 346–363.

C. Sievert. Interactive Web-Based Data Visualization with R, plotly, and shiny. Chapman and Hall/CRC Florida, 2020.

Florian Hartig (2020). DHARMa: Residual Diagnostics for Hierarchical (Multi-Level / Mixed) Regression Models. R package version 0.3.3.0. https://CRAN.R-project.org/package=DHARMa

Hunt, A., Steward, D., Richardson, M., Hinds J., Bragg, R., White, M. and Burt, J. 2017 Monitor of Engagement with the Natural Environment: developing a method to measure nature connection across the English population (adults and children) Natural England Commissioned Reports, Number 233. York.

Demographic data - Figure 3

Socioeconomic group - Figure 4

Regional information - Figure 5

Household information - Figure 6

Membership information - Figure 7

Results summary

For the purpose of this analysis participants were grouped according to whether they agreed with a statement (strongly agree and agree) compared to those who disagreed or were neutral.

Participants were more likely to agree with the statements “I always find beauty in nature” and “I always treat nature with respect” and less likely to agree with the statement “I fee part of nature” (Fig 8.1). There were differences observed between demographic groups.

Older participants were more likely to agree with the nature connection statements and this relationship was strongest for “I always treat nature with respect” (Age trend = 0.04, S.E.=0.005) and “Nature makes me happy” (Age trend = 0.03, S.E.=0.005).

Females were more likely to agree with all the nature connection statements with the exception of “I feel part of nature” (Figure 8.2).

Employed participants and homemakers were more likely to agree with the statements “Spending time in nature is important to me” and “I feel part of nature” compared to unemployed and retired participants (Figure 8.3).

White participants were significantly more likely to agree that “Nature makes me happy”, “I always treat nature with respect”, and “Time spent in nature is important” (Figure 8.4). These differences were small in magnitude; for example 97% of white participants agreed with statement “I always treat nature with respect” compared to 94% of BAME participants.

Figure 8.1

Figure 8.2

Figure 8.3

Figure 8.4

Data tables

Results Summary

For the purpose of this analysis participants who had ‘frequent’ or ‘occasional’ childhood trips to nature were grouped together and compared to those who had ‘rare’ or ‘none’. Those who said ‘Don’t know’ were removed prior to analysis.

Ethnicity was a significant predictor of childhood visits to nature in this sample. White (88%) participants were significantly more likely to report frequent or occasional childhood visits to nature compared to BAME participants (77%) (Fig 9.1).

Higher levels of education were associated with more frequent trips to nature in childhood. The significant differences between different educational groups are shown in Fig 9.2.

Participants who do not have a garden were less likely to state they had frequent or occasional trips to nature during childhood (Fig 9.3) as were male participants (Fig 9.4).

Figure 9.1

Figure 9.2

Figure 9.3

Figure 9.4

Data tables

Results summary

For the purpose of this analysis respondents who believed the number of trees in an area were ‘about right’ or ‘too many’ were combined and compared to respondents who believed there were ‘too few’ trees in an area.

Participants were more likely to endorse that there were ‘too few’ trees in towns and cities compared to neighbourhoods, and this was also true for large trees (Fig 10.1).

Gender was a significant predictor of whether participants believed there were too few trees in an area. Men tended to think there were too few trees compared to women (Fig 10.2). The difference was small in magnitude as 31% of men were estimated to believe there were too few trees in an area compared to 26% of women.

Retired individals were less likely to endorse feeling there were too few trees in an area compared to employed individuals and homemakers (Fig 10.3).

Figure 10.1

Fig 10.2

Fig 10.3

Data tables

Results summary

Overall participants found that the areas where trees made the most significant contribution were urban woodlands, parks and recreation grounds open to the public, community gardens and amenity areas (schools hospitals etc.) (Figure 11.1). These opinions did differ according to several demographics.

White participants scored the importance of trees in Urban woodlands, parks and recreation grounds, canals and riverbanks and private gardens higher than BAME participants (Figure 11.2), although the differences in scores were small (no more than half a point).

Females scored the importance of trees in most areas higher than males, except for around railway lines, roadsides and roundabouts (Fig 11.3). The differences in estimated priority scores were typically small at around half a point or less.

Participants with access to a garden believed trees in urban woodlands, canals and riverbanks and private gardens made the most significant contributions (Fig 11.4).

Although country of residence and having dependents also significantly predicted how much participants felt trees contributed to different areas after multiple testing correction there were few robust differences between these groups.

Figure 11.1

Figure 11.2

Figure 11.3

Figure 11.4

Data tables

Results summary

For the purpose of this analysis participants were grouped according to whether they were strongly supportive or supportive compared to neutral/unsupportive/strongly unsupportive.

Most participants were supportive of new tree planting across a range of different area types. People were less likely to be supportive of tree planing around railway lines, roadsides and roundabouts and private gardens, and more likely to be supportive of new tree planting in parks and recreational areas, urban woodlands and amenity areas (Figure 12.1).

Higher levels of education were associated with being more supportive of new tree planting across all areas (Fig 12.2). White participants were also more likely to be supportive of planting new trees (Fig 12.3) although this was a small effect (81% white participants vs 73% BAME participants).

Women were generally more supportive of new tree planting but this differed by area. Men were more supportive of tree planting around railway lines compared to women (Fig 12.4). Employment was also associated with support for new tree planting but different patters arose for different statements (Fig 12.5): unemployed individuals were less likely to support planting on residential streets than employed individuals and retired individuals were less likely to support planting around canals and riverbanks compared to those in employment.

Older participants were more likely to be supportive of new tree planting across a range of areas, but younger people were more likely to be supportive of planting around residential streets (Fig 12.6).

Figure 12.1

Figure 12.2

Figure 12.3

Figure 12.4

Figure 12.5

Figure 12.6

Data tables

Results summary

Overall, participants were most likely to agree with the statement, I tend to notice the trees in my neighbourhood and my town/city when I am outside" and least likely to agree with the statement, “I feel more connected to urban trees since COVID lockdown” (Fig 13.1).

Women were more likely to agree with most statements with the exception of the statement, “I am concerned that pests and diseases might be an important threat to urban trees”. Men were more likely to be concerned about pests and diseases than women (Fig 13.2).

Employment status was a significant predictor of attitudes to urban trees. Employed individuals were more likely to agree with the statement “I feel more connected to urban trees since COVID lockdown” than retired or unemployed individuals. Employed individuals were also more likely to agree with the statements “I think the trees in my neighbourhood and my town/city should be given more protection from damage and removal” than retired individuals. Employed individuals were more likely to agree with the statement “I care for trees” compared to the unemployed (Fig 13.3).

The relationship between education and attitudes to urban trees differed according to statement. The results are summarised in Figure 13.4 with individuals with higher levels of education tending more likely to agree with some statements, including “Urban trees are good for my mental and physical health”, “My town/city is a better place because of the trees” and “I care about trees”.

Ethnicity was a significant predictor of some attitudes to urban trees with white participants more likely to agree with the statement “Urban trees are good for my mental and physical health”, “My town/city is a better place because of the trees”, “I care about trees in my neighbourhood” and “I think the trees in my neighbourhood and my town/city seem to be in a healthy condition” (Fig 13.5).

The relationship with age differed according to statement. Older participants were more likely to agree with the statements in red on Figure 13.6 (“My town/city is a better place because of the trees”, “Urban trees are good for my mental and physical health” etc.). Younger people were more likely to agree with the statements in blue (“I feel more connected to urban trees since COVID lockdown”). For the statements in black there is no significant relationship between age and the proportion of people who agree with a statement.

Fig 13.1

Fig 13.2

Fig 13.3

Fig 13.4

Fig 13.5

Fig 13.6

Data tables

Analysis note

Two analyses were performed for this question. The first took the management statements; “The branches of a tree in my street being trimmed (e.g. crown reduction, pollarding)” and “Autumn leaves being cleared” and grouped respondents according to whether they were “very pleased” or “pleased” and compared them to all other respondents (neutral/annoyed/very annoyed) (summarised in ‘a’ figures).

A second analysis took the remaining respondents and grouped them according to whether they were “very annoyed” or “annoyed” and compared them to all other respondents (neutral/pleased/very pleased) (summarised in ‘b’ figures).

Results summary

Unemployed individuals were less likely to report being pleased by “The branches of a tree in my street being trimmed (e.g. crown reduction, pollarding)” and “Autumn leaves being cleared” (Fig 14.1a). Older participants were more likely to be pleased by autumn leaves being cleared and less likely to by pleased by tree branches in the street being trimmed (Fig 14.2a).

For the remaining statements participants were most likely to find “Damaged trees are left unmanaged” as the most annoying and “A neighbour removes a tree on their property” as the least annoying (Fig 14.1b). People without dependents were more likely to be annoyed than those with dependents across all statements (Fig 14.3b) although the difference was small overall (34 vs 41%). Individuals with no education were less likely to be annoyed about the remaining statements (Fig 14.2b).

Older participants were more likely to be annoyed by urban tree management practices and this was most pronounced for “Hedges in gardens and public spaces are not trimmed and grow to intrude into pavement space” and “Damaged trees left unmanaged” (Fig 14.4b).

Females were significantly more likely to report being annoyed by “Trees in my neighbourhood being given anti-wildlife measures”, “Damaged trees left unmanaged” and “Hedges in gardens and public spaces are not trimmed and grow to intrude into pavement space” (Fig 14.5b).

Retired respondents were more likely to annoyed than employed respondents by some aspects of urban tree management (Fig 14.6b). Differences according to ethnicity are shown in Figure 14.7b and by country of residence in Figure 14.8b.

Fig 14.1a

Fig 14.2a

Fig 14.1b

Fig 14.2b

Fig 14.3b

Fig 14.4b

Fig 14.5b

Fig 14.6b

Fig 14.7b

Fig 14.8b

Data tables - a

Data tables - b

Results summary

For the purpose of this analysis participants were grouped according to whether they would be very pleased/pleased about being consulted about management practices and compared to those who would be neutral/annoyed/very annoyed.

BAME participants were more likely to report being pleased if they had been consulted about urban tree management practices (Fig 15.1) as were those who have a degree or higher education (Fig 15.2). Retired or employed individuals were also more likely to report being pleased about being consulted about tree management practices (Fig 15.3) as were those with dependents (Fig 15.4).

Fig 15.1

Fig 15.2

Fig 15.3

Fig 15.4

Data tables

Results summary

For the purposes of this analysis participants were grouped so that those who agreed with the statements presented were compared to those stating ‘no’ or ‘don’t know’.

Respondents were more likely to think “…it is important for the government to set aside more money for planting new trees”, and “there should be more money set aside by the government for tree management and care”. They were less likely to agree that they would be “….willing to pay higher council tax if that meant I had a greater say in how/where the trees in my neighbourhood and my town/city are managed”. They were also less likely to think “…I am able to express my opinions about the trees in my neighbourhood and my town/city and my voice will be heard” and “I know where to find up-to-date information about policies referring to tree planting, tree health and tree management” (Fig 16.1).

Age was a significant predictor of opinion. Older participants were more likely to agree that the government should put aside money for tree planting, care and management but less likely to agree that they would pay more council tax to do so (Figure 16.2). Younger people were more likely to agree that “…I am able to express my opinions about the trees in my neighbourhood and my town/city and my voice will be heard” and “I know where to find up-to-date information about policies referring to tree planting, tree health and tree management”. They also were more likely to want to become more involved with decision making around tree management.

Participants with a degree or higher education were more likely to agree with all the statements regarding tree management (Fig 16.3) and unemployed people less likely (Fig 16.4). BAME participants were more likely to agree with the statements “…I am able to express my opinions about the trees in my neighbourhood and my town/city and my voice will be heard”,“I know where to find up-to-date information about policies referring to tree planting…” and “I would like to become more involved with decision making relating to the trees in my neighbourhood and my town/city” (Fig 16.5). People with gardens were more likely to agree with all statements (Fig 16.6.) although the difference was small (46% agree vs 42%). Participants with dependents were more likely to agree with 5/7 statements (Fig 16.7).

Fig 16.1

Fig 16.2

Fig 16.3

Fig 16.4

Fig 16.5

Fig 16.6

Fig 16.7

Data tables

Results summary

Participants were most likely to think that councils and local authorities were responsible for trees in their neighbourhoods and cites and least likely to think tree surgeons were responsible (Fig 17.1).

There was a strong positive relationship between age and the belief that councils and local authorities are responsible for local tree management - older participants were more likely to select this response. Younger participants were more likely to believe community forests and tree surgeons were responsible (Fig 17.2).

Women were more likely to believe that councils/local authorities and government agencies are responsible for trees in their neighbourhoods whereas men were more likely to believe infrastructure companies were responsible (Fig 17.3). People without dependents were more likely to believe that councils/local authorities were responsible (Fig 17.4).

Those with a degree or higher educational qualification were more likely to believe that private owners and infrastructure companies were responsible for trees in their neighbourhood (Fig 17.5). White participants were more likely to believe councils or local authorities were responsible for tree management compared to BAME participants (Fig 17.6). Differences according to employment status are shown in Figure 17.7.

Fig 17.1

Fig 17.2

Fig 17.3

Fig 17.4

Fig 17.5

Fig 17.6

Fig 17.7

Data tables

Results summary

Participants were less likely to believe tree surgeons and tenants should receive more financial support to manage trees in their neighbourhood (Fig 18.1). Older participants were more likely to believe that councils and local authorities and government agencies should receive more funding (Fig 18.2).

There were small but significant differences in beliefs about financial support between various different demographic groups. Women were more likely to believe that community forests and environmental organisations should receive more funding (Fig 18.3). Participants with children were more likely to believe infrastructure companies and private owners should receive additional funding (Fig 18.4).

White participants were more likely to believe councils and local authorities should receive more funding whereas BAME participants thought more funding should go to private owners and infrastructure companies (Fig 18.5). Few robust differences between employment groups were observed.

Figure 18.1

Figure 18.2

Figure 18.3

Figure 18.4

Figure 18.5

Data tables

Results summary

The most commonly endorsed activity was “Planted a tree/s in my garden” (Fig 19.1). Younger participants were more likely to have done the majority of the listed activities with the exception of “Carried out maintenance myself on a tree I own” (Fig 19.2).

Participants with dependents were more likely to have performed the majority of activities listed (Fig 19.3). Those with higher education were more likely to have planted trees in their garden, employed a tree surgeon on their own property and carried out maintenance on a tree they own (Fig 19.4).

There were differences in the proportion of participants selecting an activity based on ethnicity. White participants were less likely to report protesting the removal of a tree, lobbied for tree removal or complained about a tree in their neighbourhood compared to BAME participants. BAME individuals were more likely to report having spoken to a neighbour about a tree on their property (Fig 19.5).

Retired individuals tended to endorse more tree management activities (Fig 19.6). People residing Wales tended to endorse less (Fig 19.7), particulary when compared to those residing in England. Garden owners also endorsed more of these activities (Fig 19.8).

Fig 19.1

Fig 19.2

Fig 19.3

Fig 19.4

Fig 19.5

Fig 19.6

Fig 19.7

Fig 19.8

Data tables

Results summary

Participants were most likely to state they were interested in ‘none of the above’ or ‘Planting and manage own trees in your garden’ (Fig 20.1). Older participants were more likely to state they were not interested in the listed activities and younger participants were more likely to state they would be interested in donating money (Fig 20.2).

Women were more likely to be interested in planting a tree in their own garden (24% vs 20%) and men were slightly more likely to be interested in becoming a tree warden (6% vs 5%) (Fig 20.3).

Higher education was associated with increased interest in the majority of listed activities (Fig 20.4). Employed individuals were more likely to volunteer to water trees in public spaces, donate money, and join community groups. Retired individuals were less likely to be interested in planting trees in public spaces (Fig 20.5). White participants were less likely to be interested in donating money and volunteering to water trees in public places than BAME participants (Fig 20.6).

People residing in England were more likely to be interested in joining events and campaigns related to trees compared to those residing in Wales (Fig 20.7). Individuals with gardens were more likely to be interested in planting a tree in their own garden (26% 95% C.I. 12-29% vs 18% 95% C.I. 15-21%).

Fig 20.1

Fig 20.2

Fig 20.3

Fig 20.4

Fig 20.5

Fig 20.6

Fig 20.7

Data tables

Results summary

Television, printed media, the websites of national organisations and local authority websites were the forms of communcation found most effective across all participants (Fig 21.1).

Older participants were more likely find local authority websites and national newspapers effective and younger participants more likely to find YouTube/Instagram/Vlogs, radio, Twitter, TV and printed media (Fig 21.2).

Individuals with dependents were more likely to find YouTube/Instagram/Vlogs and radio effective forms of communciation (Fig 21.3). Higher levels of education were associated with finding several forms of communcation effective (Fig 21.4). BAME individuals were more likely to find YouTube/Instgram/Vlogs, printed media and radio as effective (Fig 21.5). There were differences according to employment status with retired individuals less likely to find Twitter effective and more likely to find radio useful (Fig 21.6).

Fig 21.1

Fig 21.2

Fig 21.3

Fig 21.4

Fig 21.5

Fig 21.6

Data tables

Results summary

Participants were more likely to agree that “Polluting industries/sectors should help to pay for tree management and planting in neighbourhoods and towns/cities” and “Central government should make more money available to maintain and enhance tree cover in towns and cities across the UK” (Fig 22.1).

The relationship between age and the proportion of people who agreed with a statement differed by statement. Older participants were more likely to believe “Polluting industries/sectors should help to pay for tree management…” and “Central government should make more money available…”, whereas younger participants were more likely to “…already support urban trees through appropriate charitable donations” and agree with the remaining statements (Fig 22.2).

Women were more likely to think “Polluting industries/sectors should help to pay for tree management…”, whereas men were more likely to already support tree charities, believe “Those within towns and cities with better access to trees should pay more in council tax than those with poorer access” and be “…willing to pay higher council tax if I knew the additional sum would be dedicated to tree care and planting” (Fig 22.3).

Those with a degree or higher education were more likely to be “…willing to pay higher council tax…”, “…already support urban trees through appropriate charitable donations” and would support ‘adopt a tree’ schemes (Fig 22.4).

Employment category also predicted opinions: retired and employed individuals were more likely to be “…willing to pay higher council tax…” and employed individuals more supportive of ‘adopt a tree’ schemes (Fig 22.5).

People with dependents were more likely to agree with a multitude of statements (Fig 22.6).

BAME participants were more likely to already support tree charities and believe that “Those within towns and cities with better access to trees should pay more in council tax than those with poorer access” than white participants (Fig 22.7).

Figure 22.1

Figure 22.2

Figure 22.3

Figure 22.4

Figure 22.5

Figure 22.6

Figure 22.7

Data tables

Results summary

As the weighted nature connection index score is a continuous variable the figures show the estimated proportions of participants agreeing with a particular statement/question at various values of the score. Hunt et al. (2017) split the participants in their sample into quintiles according to the value of the nature connection score and describe these quintiles as low/medium-low/medium/medium-high and high connection to nature. The mid-points of the score quintiles in the present sample are 29, 62, 70, 81 and 87 and these are plotted on the figures and described as low/medium-low/medium/medium-high and high nature connection scores respectively.

Participants with a higher nature connection index score were more likely to feel there were too few trees in neighbourhoods and town/cities (Fig 23.2), were more supportive of tree planting in all areas (Fig 23.3) and were more likely to agree with all the statements describing the benefits of urban trees (see Q15, Fig 23.4).

Fig 23.2

Fig 23.3

Fig 23.4

Data descriptives

Roughly thirty percent of participants scored “Absorbing carbon dioxide and supplying oxygen”, “A habitat and shelter for wildlife” and “Cleaner, less polluted air” as ‘10’, the most important benefit of urban trees. Only 8% of participants believed the most important benefit of urban trees was to increase property value.

Data descriptives

The disbenefit of urban trees most commonly scored as ‘most important’ (score of 10) was the blocking of urban light (8.2% of participants). The disbenefit most commonly ranked as least important (score of 1) was the reduction of property value (9.2% of participants).

Data descriptives

Twenty-eight percent of participants ranked ‘Urban woodlands or forest’ as the most important urban habitat to protect. One third of participants ranked ‘Wasteland/brownfield’ as the least important urban habitat to protect.