The Impact Of Children's Connection To Nature

11Table 1: Table showing the % of pupils eligible for free schools meals andwho speak English as their first language in each school.School 1School 2School 3School 4School 5School 6School 7School 8School 9School 10School 11School 12School 13School 14School 15Free SchoolMeals sh as FirstLanguage cipants fromSchool40 (5.2%)53 (6.8%)93 (12%)32 (4.1%)57 (7.4%)74 (9.5%)27 (3.5%)50 (6.5%)40 (5.2%)56 (7.2%)23 (3.0%)81 (10.4%)27 (3.5%)46 (5.9%)76 (9.8%).Data accessed via: dtheir-characteristics-january-2013ParticipantsIn total, 775 children aged 10-11 years responded to the questionnaire andwere all in year 6 in primary school. Four hundred and seven of these weremale (52.5%) and 368 were female (47.5%). In the demographics section ofthe questionnaire, children could identify with one of six ethnic groupsprovided. Of the children who took part, 615 identified themselves as white(79.4%); 44 children as mixed (5.7%); 68 children as Asian or Asian British(8.8%); 17 children as Black or Black British (2.2%); 14 children identified withother ethnic group (1.8%) and 16 children’s ethnic group was not stated(2.1%). The sample size of the research (n 775) can be regarded assufficient and acceptable for the implementation of a quantitative measure, asTinsley and Tinsley (1987) suggested a ratio of 5 to 10 subjects per item.With a total of 38 items, a sample size in excess of 380 is sufficient.Ethical ConsiderationsThe research was cleared through the University of Derby PsychologyResearch Ethics Committee and adhered to the British Psychological Societyethical guidelines. In order to obtain participation, consent letters were senthome to parents through the school, outlining the research being conducted incollaboration with the RSPB and giving them the opportunity to request thattheir child did not complete the questionnaire. The consent letter alsoinformed parents of the need to obtain children’s attendance and teacherassessments for Mathematics and English, although this would be keptconfidential. The child’s right to withdraw their data 1 month after datacompletion was also outlined.www.derby.ac.uk/

12RecruitmentPrior to data collection, permission to conduct the research was sought fromhead teachers of schools or the head of year six, through a phone call andemail which outlined the rational and requirements of the research. This alsoreassured the schools that the research had received ethical clearance andthat children’s questionnaire responses, attendance and teacher assessmentlevels would be kept anonymous. Once a school had expressed interest inthe research, and had given initial consent to participate, they were informedthat the school would receive a number of gifts from the RSPB for taking part.In addition, a date was arranged for consent forms to be taken and left withthe school reception, to be distributed to year 6 children to take home to theirparents for their consideration. A response time frame of one week wasimplemented and stipulated on the consent letter. After this period hadelapsed, the researcher returned to the schools to collect any returned opt-outslips. Following this, a date and time for data collection was arranged witheach school to suit their convenience.ProcedureThe questionnaires were handed out to each year 6 class in register order.The attendance for each child and their teacher assessments for English andMathematics were also requested in register order for each class, so thateach child’s information could be matched to their questionnaire, without theneed for names. This ensured anonymity. The questionnaires werenumbered to allow the researcher to omit the relevant attendance and teacherassessments, in the case of a child being absent. When handing out thequestionnaires, the researcher noted the class identity code used in theschool, i.e. 6SH, again, so that this could be accurately matched toattendance and teacher assessments.Prior to the children completing the questionnaire, the researcher wasintroduced and briefly outlined the questionnaire and what the children wouldneed to do. Children were told that their parents had given consent for themto take part and were informed of their right to stop at any stage through thecompletion of the questionnaire. They were further assured that theirresponses were confidential and that there were no wrong answers, and thusnot a test in any way. Collectively as a class, children were guided throughthe process of completing their unique identification code and were told thatthis was how they would be identified, rather than providing their name.All children who participated in the research completed the questionnaire in aclassroom setting. Although children could start the questionnaire, somerequired assistance with their ethnic group choice and with the items whichgauged how many days they had spent outdoors and in nature during the pastweek. For the most part, the year 6 children who participated were able toread and comprehend the questions without any problems, although somesought clarification on what a question was asking them. In some schools,children worked through the questionnaire with support from a teachingassistant, although their responses were their own. Once all children in awww.derby.ac.uk/

13class had completed the questionnaire, these were collected in by theresearcher and the children were thanked. They were then provided with aresearch debrief with informed them that the questionnaires were for theRSPB who were looking at the relationship between children’s engagementwith natural world and their well-being, pro-nature behaviour and educationalattainment. The children were also informed that due to their participation,they had earned some rewards for their school on behalf of the RSPB.Data EntryThe data gathered by the questionnaires was input onto SPSS.Quality AssuranceIn order to ensure that the data had been input correctly, and that any missingvalues were not input errors, 20 questionnaires were drawn at random andwere checked against the information previously entered on SPSS.AnalysisThe analysis of the data explored the internal reliability of the scores for theConnection to Nature Index, Children’s Pro-environmental Behaviour Measureand the Life Satisfaction Scale. The response distribution of each of the foursub-scales defined above was analysed and did not impact on the reportedfindings. Correlations were explored between subscales, English andMathematics scores, attendance and days spent outdoors and in nature overthe past week. In addition, multiple regression and mediation analysis wasconducted to determine predictors of English attainment further. Curvilinearanalysis revealed linear relationships, therefore the existing benefit thresholdlevel for CNI (i.e. 1.5) was used as the basis for tests of difference above andbelow the threshold.www.derby.ac.uk/

14ResultsDemographicsTable 2 presents demographic data including gender, age and ethnicity ofchildren for the overall sample size of year 6 children.Table 2: Summary of the participant demographicsDemographicNumber of ChildrenPercentage ofChildrenTotal Number(%)Overall 7.5%Aged 10 years24031%774(99.9%)Aged 11 years53468.9%White61579.4%Mixed445.7%Asian or AsianBritish688.8%Black or BlackBritish172.2%Other EthnicGroup141.8%Not Stated162.1%774(99.9%)In table 2, it can be seen that more males participated in the questionnairesurvey over the 15 schools; females accounted for 5% less of the participantwww.derby.ac.uk/

15total than males. In terms of age, children aged 11 years were morerepresented than children aged 10 years, with a difference of 294 participants.However, age 11 is the more typical age for the year 6 children from Januaryonwards. The results for ethnicity show that the majority of children in thisresearch identified themselves as white, accounting for 79.4% of ethnicity.The next highest represented ethnicity was Asian or Asian British, with 8.8%of children identifying with this group, although was significantly less thanwhite respondents. Children who identified as mixed accounted for 5.7% ofthe total sample size, whilst Black or Black British, other ethnic group and notstated were accounted for by a similar number of children, and were the leastrepresented ethnic groups. A comparison of CNI scores across demographicgroups are shown in table 3.Table 3 - The mean CNI and standard deviations by demographicsDemographicsMeansStandard .95.60Asian or Asian British0.98.50Black or Black British0.73.52Other1.34.26Not Stated0.79.61Analysis of connection to nature and gender showed that females were moreconnected to nature than males (t -7.44, df 761.55, p 0.01). Samplesizes did not allow any meaningful investigation of differences in relation toethnicity.Internal Reliability of the MeasuresThe 16-item Connection to Nature Index (Cheng & Monroe, 2012) was foundto have a high internal reliability score after implementation with 775 year 6children (α .84). This is a similar value to that obtained in previous research(α .87). In addition, the 5-item pro-environmental behaviour measure (α .74) and the 7-item Life Satisfaction Scale (α .80) were both found to havehigh internal reliability.www.derby.ac.uk/

16Connection to Nature ScoresThe children’s overall CNI score, and score on the four sub-scales wascalculated. The results are presented in table 4.Table 4 - The mean score for the overall CNI scale and each of the four subscalesMeanStd.DeviationOverallCNICNI EnjoymentCNI EmpathyCNI OnenessCNI ests of Relationships with CNIBivariate correlations were conducted in order to investigate whether therewere any associations amongst the sub-scales measures (connection tonature; pro-environmental behaviours; pro-nature behaviours; life satisfactionand; health) of the questionnaire, teacher assessments for mathematics andEnglish, attendance and the number of days spent outdoors and in natureover the last week. See Appendix B for the full correlation matrix. Selectedresults are listed by measure below.Connection to Nature IndexThe results demonstrated that there was a weak correlation between the CNIscores and the 1-item health measure (Pearson’s r(768) .09, p .01) andthe CNI and mean scores of the Life Satisfaction Measure (Pearson’s r(775) .14, p .01).A positive correlation was also evident between the CNI and days spentoutdoors over the past week (Pearson’s r(775) .17, p .01) and a strongercorrelation between the CNI and the number of days spent in nature over thepast week (Pearson’s r(771) .30, p .01), suggesting that the more timespent in nature is associated with a child’s connection to nature.However, the strongest correlations were found between the CNI and the totalscores for the pro-nature behaviours (Pearson’s r(775) .59, p .01) and theCNI and the pro-environmental behaviour mean scores (Pearson’s r(775) .60, p .01).In contrast, no correlations were evident between the CNI scores and teacherassessments for English (Pearson’s r(775) .03, p .456) and Mathematics(Pearson’s r(775) .001, p .98). Furthermore, a correlation was not foundbetween CNI scores and children’s attendance (Pearson’s r(775) -.08, p .026).www.derby.ac.uk/

17The correlation graph shows that high scores on the Connection to NatureIndex, which indicate an agreement with the positive affect statementstowards nature, have a positive association with high scores on the proenvironmental behaviour measure, which also indicate positive behavioursrelating to supporting the environment.Regression Analysis: Connection to Nature IndexUsing regression analysis with CNI as the dependent variable, it was foundthat life satisfaction mean scores, nature behaviours total scores, proenvironmental behaviour mean scores and days spent in nature over the lastweek explained a significant amount of the variance of connection to nature(F(4, 766) 170.01, p .01, R2 .47, R2Adjusted .47). This demonstrates thatapproximately 47% of connection to nature can be accounted for by thesevariables.Health MeasureThe 1-item health measure was found to have a correlation with the meanscores of the Life Satisfaction Measure (Pearson’s r(768) .26, p .01), themean scores of the pro-environmental behaviours measure (Pearson’s r(768) .11, p .01) and the total scores of the pro-nature behaviours measure(Pearson’s r(768) .14, p .01).The health measure was also found to correlate with the number of daysspent outdoors over the past week (Pearson’s r(768) .14, p .01) and thenumber of days spent in nature over the past week (Pearson’s r(764) .11, p .01).Pro-environmental Behaviour MeasureThe mean scores of the pro-environmental behaviour measure were found tocorrelate with total scores of the pro-nature behaviours measure (Pearson’sr(775) .56, p .01). Moreover, the environmental behaviour measure wasfound to correlate with the number of days spent outside over the past week(Pearson’s r(775) .16, p .01) and the number of days spent in nature overthe past week (Pearson’s r(771) .28, p .01).Pro-nature BehavioursThe total scores of the pro-nature behaviours measure were found tocorrelate with the number of days spent outdoors over the past week(Pearson’s r(775) .17, p .01) and the number of days spent in natureover the past week (Pearson’s r(771) .25, p .01).Additional Education Attainment CorrelationsWhen refined Teacher Assessments for English were explored, i.e. using thedata in which more refined SATs levels (e.g. 4a, 4b or 4c, rather than 4) wereobtained (n 512) significant but small correlations were found betweenEnglish attainment and attendance (Pearson’s r(512) .092, p .04).www.derby.ac.uk/

18Significant but small correlations were also found between English attainmentand connection to nature (Pearson’s r(512) .10, p .02) and mean scores oflife satisfaction and English attainment (Pearson’s r(551) .18, p .01).Although weak, this suggests that connection to nature is as important tochildren’s achievement in English as life satisfaction and attendance.Regression Analysis: English AttainmentAs above, this regression analysis only included the participants who had hadmore refined English attainment provided (n 512). Using regressionanalysis with English attainment as the dependent variable, it was found thatthe Connection to Nature Index, Attendance and Life Satisfaction werepredictors of English attainment (F(3, 508) 6.07, p .01, R2 .04, R2Adjusted .03). However, only 4% of the variance of English attainment can beaccounted for by the predictor variables.Mediation AnalysisThe correlations indicated potential mediation within the relationship betweenCNI and life satisfaction. A mediation analysis was conducted to furtherexplore the relationship between connection to nature and English attainment,with life satisfaction as a mediator. There was a significant total ef

Connecting with nature should be part of every child's life as it has the potential to aid nature's revival while benefiting the child. To embed nature connection within our social norms, there is a need to be able to understand the benefits and set targets for levels of nature connection.