1. BACKGROUND AND AIMS There is a gap iŶ the edžploratioŶ of ĐhildreŶs uŶderstaŶdiŶg of their ďehaǀiour ǁheŶ theLJ are in their early years at school. Past research has shown that: • There are impacts of different peer behaviours, particularly in aggression; pro-social ďehaǀiour aŶd ǁithdraǁŶ ďehaǀiour, oŶ ĐhildreŶs outĐoŵes. I.e. LiŶks ďetǁeeŶ aggressioŶ and peer acceptance/rejection; protective factors or pro-social behaviour against aggression and peer rejection; and a reduction in quality of peer relationships as a result of withdrawn behaviour. • Behaviour plays a large role in the early school years. The prevalence of aggression and pro- social behaviour changes at different points in the early school years. I.e. an increase in verbal aggression and decrease in physical aggression as children get older; and age differences in the development of pro-social behaviour. In contrast, withdrawn behaviour has been found to remain stable from age two across contexts (Rubin, Coplan & Bowler, 2009). • Self reports can be accurate . I.e. Ladd and Kochenderfer-Ladd (2002) found that younger children made more accurate reports of victimisation than their peers. • There are few behavioural measures which use self reports from young children. • The majority of research which compares cross-informant reports of behaviour has compared self, peers and teachers and focused on children aged six and over. • Children also think that Teaching Assistants play an important part in the school community (e.g. Fraser and Meadows, 2007), but have not been included in cross-informant studies. This poster presents results from phase 1 of a 15 month longitudinal study, exploring eleven types of behaviour . There are two main research questions: 1. What is the general trend of discrepancy in behaviour reports made by children about themselves, by their peers and by teaching staff? 2. Is there a difference in the discrepancy in behaviour reports made by children about themselves, by their peers and by teaching staff in Reception (ages 4/5) and Year 1 (ages 5/6)? 2. METHODS Design: An independent groups design was used to compare cross-informant behaviour ratings and report discrepancies in Reception and Year 1. Independent variables: Research q1 – informant; Research q2 – year group Dependent variables: Research q1 – ratings; Research q2: Discrepancy in ratings Participants: 285 children across 5 schools (see diagram to right); 10 Class Teachers and 18 Teaching Assistants Materials: • 11 videos with stick people showing a range of behaviours (Rix, 2012) were presented on a laptop. • Paper-based matrix describing each of the 11 behaviours, a list of children in each class, columns to select ratings of lots; soŵetiŵes; or Ŷeǀer, for Teachers and Teaching Assistants Measures included the following behaviours: • Sharing, caring and group inclusion (pro-social) • Physical, verbal, direct relational, and indirect relational aggression; ringleader of physical and verbal aggression • Passive and active withdrawal Procedure – Children’s responses (self and peer reports): 1 to 1, 30 minute session using the following procedure for each of the behaviours Procedure – Teacher and Teaching Assistant responses (teaching staff reports): Teachers and Teaching Assistants were asked to complete the behaviour matrix based on each child, on the day the task was carried out with them or as close to this as possible. 3. RESULTS AND DISCUSSION Research question 1: Mean ratings were calculated across teaching staff and peers and compared to self reports across behaviours for each child. High ratings presented higher frequency of behaviour. Table 1 shows the trend of ratings, developed through comparing descriptive statistics. Table 1 shows that: 1. The highest ratings of pro-soĐial ďehaǀiour aŶd ǁithdraǁŶ ďehaǀiour Đaŵe froŵ self reports. ChildreŶs peers provided lower ratings than teaching staff for these behaviours. 2. The highest ratings of aggression came from teaching staff. Peer ratings mostly aligned with self reports for aggression. Research question 2: A discrepancy score was created for each child, through calculating the difference between informant ratings. Mann Whitney tests were used to compare discrepancy scores between Reception and Year 1. Overall, there were no significant findings related to discrepancies between reports for withdrawn behaviour. The following bar charts show the mean discrepancy scores where there were significant differences between Reception and Year 1 for reports aggression and pro-social behaviour: 4.CONCLUSIONS AND NEXT STEPS In conclusion, there are different patterns of agreement/disagreement between children and other informants for different behaviour types. Next steps will involve: • Further statistical comparison of discrepancy scores for each behaviour type. • Explore ratings and discrepancy scores by gender and age, as well as year group • Explore how reports change over the course of a school year, as part of phase 2 and 3 of this research. 5. REFERENCES Fraser, C., & Meadoǁs, S. ;Ϯ008Ϳ. ChildreŶs ǀieǁs of TeaĐhiŶg AssistaŶts iŶ priŵarLJ sĐhools. Education 3-13, 36(4), 351- 363. doi:10.1080/03004270701754219 Ladd, G. W., & Kochenderfer-Ladd, B. (2002). Identifying victims of peer aggression from early to middle childhood: Analysis of cross- informant data for concordance, estimation of relational adjustment, prevalence of victimization, and characteristics of identified victims. Psychological Assessment, 14(1), 74-96. doi:10.1037//1040-3590.14.1.74 Rix, K. (2012, July). Exploring self perceptions of behaviour towards peers in children aged four to seven. Presentation at PsyPAG Annual Conference, Newcastle, United Kingdom Rubin, K. H., Coplan, R. J., & Bowker, J. C. (2009). Social withdrawal in childhood. Annual review of psychology, 60(c), 141- 71. doi:10.1146/annurev.psych.60.110707.163642 Exploring children’s understanding of their behaviour in their first two years at school: with a focus on aggression, pro-social and withdrawn behaviours Katie Rix, Pam Maras & Claire P Monks Department of Psychology and Counselling, School of Health & Social Care, University of Greenwich 5 x schools 285 children Reception class 141 children Mean age: 4.64 Male: 66 Female: 75 Year 1 class 144 children Mean age: 5.58 Male: 77 Female: 67 Video Shown Asked ǁhats happening in video Confirmed/ explained by experimenter Can they think of anyone in class who behaves in this way (lots / sometimes) Do they ever behave in this way? (lots/ sometimes/ never – via response pad) Table 1: Trend of behaviour ratings across informants BPS Developmental Section Annual Conference, 2012 For more information please contact: K.R.Rix@greenwich.ac.uk -2 = Complete discrepancy Self< peers/teaching staff Peers< Teaching staff 0 = No discrepancy +2 = Complete discrepancy Self> peers/teaching staff peers> teaching staff NB: This study was conducted as part of longitudinal research for which children are met at three points in time between April 2012 and July 2013. In addition, this study was part of wider research where children are also asked about their reasons for their behaviour. 1.47 1.49 0.18 -0.02 -0.02 -0.03 1.42 1.62 0.01 0.03 0.00 0.00 -0.8 -0.4 0 0.4 0.8 1.2 1.6 2 Caring Group inclusion Direct relational aggression Physical aggression Ringleader of physical aggression Verbal aggression Mean Discrepancy Score Reception Year 1 0.29 0.5 -0.3 -0.17 -0.32 -0.22 0.15 0.32 -0.5 -0.44 -0.44 -0.39 -0.8 -0.4 0 0.4 0.8 1.2 1.6 2 Sharing Caring Direct relational aggression Indirect relational Verbal aggression Ringleader of verbal aggression Mean Discrepancy Score Reception Year 1 -0.5 -0.52 -0.3 -0.13 -0.29 -0.24 -0.68 -0.67 -0.49 -0.13 -0.5 -0.39 -0.8 -0.4 0 0.4 0.8 1.2 1.6 2 Group inclusion Direct relational aggression Indirect relational Ringleader of physical aggression Verbal aggression Ringleader of verbal aggression Mean Discrepancy Score Reception Year 1 Figure 1: Self and peer discrepancies (**p<0.01; *p<0.05) Figure 2:Self and teaching staff discrepancies (**p<0.01; *p<0.05) Figure 3: Children’s peers and teaching staff reports (**p<0.01; *p<0.05) U = 7422, N1=141, N2=144 ** U = 8250, N1=141,N2=144** U = 7109, N1=141, N2=144 ** U = 8725 N1=140, N2=144 * U = 8308, N1=140, N2=144 ** U = 8621, N1=141, N2=144 * U = 8574, N1=140, N2=144 * U = 8387 N1=140, N2=144 * U = 8655, N1=140, N2=144 * U = 8101, N1=140, N2=144 ** U = 8673, N1=140, N2=144 * U = 8281, N1=139, N2=144 ** U = 8023, N1=140, N2=144 ** U = 8305, N1=140, N2=144 ** U = 7667, N1=140, N2=144** U = 7499, N1=140, N2=144 ** U = 7325, N1=140, N2=144 ** U = 8265, N1=140, N2=144 ** Figure 1 shows that self and peer discrepancy scores were: 1. Lower for reports of caring and most aggressive behaviour in Year 1 2. Higher for reports of group inclusion in Year 1 Figure 2 shows that self and teaching staff discrepancy scores were: 1. Higher for reports of aggressive behaviour in Year 1 2. Lower for reports of sharing and caring in Year 1. Figure ϯ shoǁs that ĐhildreŶs peers aŶd teaĐhiŶg staff disĐrepaŶĐLJ sĐores ǁere: 1. Higher for reports of group inclusion and aggression in Year 1. Highest ratings pro-social behaviours Aggression Withdrawn behaviours Self Teachers Self Teaching staff Peers Self Teaching staff Peers Peers