Positive Transition to Middle School: A Multilevel Model Analysis of a Portuguese School Adjustment Program

Abstract

Transition to middle school in Portugal is associated with increased levels of stress for students, as well as an increase in grade retention. This study investigated the impact of a school adjustment program designed to support students during this transition, and its differential results by gender and fourth-grade class size. Multilevel linear modeling (MLM) with a repeated-measures design was used. Three cohorts of students (n = 743) completed the School Stress Survey in fourth (middle and end) and fifth grades (end). In addition, fifth-grade school records were collected for information about absenteeism and transition. After adjusting for gender and fourth-grade class size, positive interaction effects from the intervention were found in academic and teacher/rules stress, as well as higher rates of transition to sixth grade and lower rates of absenteeism. Differential effects by gender also were found with girls displaying higher school success and lower school absenteeism along the transition; however, they also displayed larger increases in academic and peer-related stress during the period. No gender differences in benefits from program participation were found. The positive results support the implementation of the program among Portuguese fourth graders.

Keywords

school stress school absenteeism school success middle school transition school adjustment program

Introduction

School transition is a life-changing event for children (Symonds, 2015) and, as such, periods of transition are often stressful. As Lane, Oakes, Carter, and Messenger (2015) suggested, schooling constitutes a series of overlapping transitions for all students, with middle school transition requiring that students adapt simultaneously to new school environments, different social interactions, and greater academic expectations (Anderson, Jacobs, Schramm, & Splittberger, 2000). In addition, in their first school transitions, due to their early age, students have few previous transition experiences (Arens, Yeung, Craven, Watermann, & Hasselhorn, 2013; Symonds & Galton, 2014), and those who do not successfully navigate this transition are at heightened risk for future school failure (Lane et al., 2015).

Even though entry into middle school marks a period of potential change and adjustment (Lane et al., 2015), negative effects of middle school transition have been widely studied since the 1980s (Anderson et al., 2000; Eccles & Roeser, 2009). Eccles and Roeser (2009) concluded that there is substantial evidence of declines in academic motivation and achievement in middle school transition, whereas Rice, Frederickson, and Seymour (2011) described negative impacts from this transition on well-being, school engagement, and psychological adjustment. Also, several studies (Akos & Galassi, 2004b; Qualter, Whiteley, Hutchinson, & Pope, 2007; Symonds & Galton, 2014) concluded that most students report some concerns and anxieties prior to transition, the most common being bullying, getting lost, making new friends, increased workload, and dealing with new environments and routines. International data are also consistent in identifying a decrease in grades following transfer to middle school, even if it occurs at different ages in different educational systems (Akos & Galassi, 2004b; Arens et al., 2013; Authors, 2016; Pietarinen, 2000).

The Portuguese mandatory school system is divided into four cycles: elementary school (first to fourth grades), marked by a single teacher (teaching Portuguese, Mathematic, Sciences, and Arts), and smaller schools that are closer to the communities; low middle school (fifth and sixth grades) characterized by larger schools and one teacher per each of the nine subjects; high middle school (seventh to ninth grades), usually housed in the same schools as the previous cycle but encompassing completely different teachers and additional subjects; and secondary school (10th–12th grades), where students can choose from four different areas of studies (sciences, economics, arts, and social sciences) or courses more oriented toward an earlier entry to the labor market (chosen by 40% of students, Direção Geral de Estatísticas da Educação e da Ciência [DGEEC], 2015).

Stress in Middle School Transition

Stress is present in everyday life of children and adolescents, and it may bear negative consequences for their physical, cognitive, and socioemotional well-being (Lazarus, 2000). Psychological stress “is a relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being” (Lazarus & Folkman, 1984, p. 21). Long-term exposure to stress is associated with maladaptive emotional or social adjustment (Lazarus, 2000). It is impossible to eliminate every source of stress from the lives of children, and so it is desirable that they deal with it effectively, in ways that reduce negative consequences on their health and well-being.

Research has determined that middle school transition is associated with stress. For example, Goldstein, Boxer, and Rudolph (2015) found that middle school transition stress was negatively associated with academic performance. Coelho and Romão (2016) concluded that fifth graders (first year after transition) had higher levels of academic stress and teacher/rules stress than fourth graders (before transition). In addition, most students who experienced high stress levels manifested poorer attendance and behavioral problems (Coelho & Romão, 2016).

Three domains of school life have been identified as the primary areas of stress for students during middle school transition (Akos & Galassi, 2004b), namely, academic, teacher relations/school rules, and peer relations. Academic concerns such as increased homework, pressure to do well, and potential drops in achievement are common for students and parents (Akos & Galassi, 2004b). Coelho and Romão (2016) considered the increase in homework (with different subjects and additional tests) as a potential academic stressor because, in Portugal, the curriculum is fragmented into a considerable number of subjects (nine), whereas there are also changes in grading practices (grades go from qualitative to quantitative). These changes require good organization and study management skills which students usually do not possess (Pereira & Mendonça, 2005).

There is also added pressure with the teachers and school rules, as there are more rules by which to abide. In middle school, children are asked to be more autonomous in managing their increasingly more complex schedules, finding classrooms, and bringing the appropriate materials to class. In Portuguese elementary schools, a single teacher is responsible for teaching all subjects, whereas in middle school, students must learn to navigate the expectation of eight (or more) teachers, bringing potentially more conflict in student–teacher relationships. Students need to adapt to teachers with different teaching styles and demands and who spend much less time with them than elementary school teachers (Coelho & Figueira, 2011). Another source of stress is that teachers place a greater emphasis on control and discipline while demanding more autonomous behaviors (Wenz-Gross, Siperstein, Untch, & Widaman, 1997).

Studies carried out in Finland and Portugal (Pereira & Mendonça, 2005; Pietarinen, 2000) concluded that peer-related stress was the most pervasive dimension of stress during middle school transition. In Portugal, middle schools are larger schools, where students from several feeder elementary schools converge (even though students remain within the same school grouping), thereby disrupting established peer networks. Several authors (Authors et al., 2016; Pereira & Mendonça, 2005; Pietarinen, 2000) identified a range of concerns regarding peer relations including not being accepted by classmates, unpleasant initiation rituals, loss of status, and the worries that accompany this loss, whereas Arens et al. (2013) found that peer support had a protective effect during the transition period.

Thus, transitions frequently cause disturbances in the adaptive functioning of children, requiring that parents, teachers, and children learn to manage stress levels. Goldstein et al. (2015) suggested that students would benefit from efforts to reduce transition stress. However, in Portugal, most students are not included in formal programs to prepare for this transition (Coelho & Romão, 2016), even though this transition takes place at an earlier age (around 9 years) when compared with other countries (e.g., the United Kingdom or Spain, where this transition takes place at age 11), which has substantial implications for students’ academic success and personal well-being.

Individual differences in middle school transition

Although school transitions are sources of stress for all students, it is widely acknowledged that there are individual differences in adjustment, making some students more vulnerable than others to school disengagement during middle school transition (Qualter et al., 2007; Rice et al., 2011; Ryan, Shim, & Makara, 2013). Among researchers, age gathers some consensus as a factor that influences this transition, with younger students reporting more problems with the transition (Anderson et al., 2000; Arens et al., 2013; Authors, 2016). Arens et al. (2013) reported that making the transition at age 9 led to lower self-esteem, academic competence perceptions, and academic affect. Therefore, there is additional need for studies analyzing middle school transition in countries with earlier middle school transitions, such as Portugal.

Gender may also be a salient factor to consider in middle school transition (Lane et al., 2015), even though the literature is not entirely consistent, several authors have reported no gender differences in this transition (Cotterell, 1992; Goldstein et al., 2013; Ryan et al., 2013), whereas other authors found that girls were more vulnerable (Anderson et al., 2000; Chung, Elias, & Schneider, 1998), especially to peer-related stress (Coelho and Romão, 2016; Chung et al., 1998). Chung et al. (1998) reported that after middle school transition, girls found peer relationships more stressful. Similarly, Coelho & Romão (2016) found that girls had a greater increase of peer-related stress with this transition. Conversely, there are authors who reported greater declines in academic achievement for boys (Akos & Galassi, 2004a; Pereira & Mendonça, 2005) or concluded that boys displayed higher levels of teacher/rules stress after the transition (Coelho & Romão, 2016), as well as higher levels of disciplinary incidents and grade retention rates (Pereira & Mendonça, 2005).

Group differences in middle school transition

Cotterell (1992) concluded that children moving to larger schools were more anxious than those transferring between similarly sized schools, whereas Symonds (2015) found that children are less stressed if the environmental differences between schools are small. Symonds concluded that having more friends made children feel less vulnerable in the larger and unfamiliar peer group after transition and that entering a new school environment is a major challenge for children best understood and supported by others who share the same experience. Therefore, students from larger classes, who have a larger number of friends sharing the transition, should report smaller increases in stress with middle school transitions.

Interventions to reduce stress in middle school transition

Given the importance of middle school transition for future adjustment and educational success, it is crucial to develop programs to help navigate the transition (Coelho & Romão, 2016; Rice et al., 2011; Symonds & Galton, 2014). Symonds (2015) proposed that transition interventions should support well-being and adjustment by addressing both children’s personal and social skills and their relationships. The author suggested that interventions should cover several areas of well-being, including identity, self-esteem, competence, autonomy, and social support. Regarding stress, Bloyce and Frederickson (2012) reported that in the United Kingdom, the most anxious students who participated in a six-session program showed greater decreases in concerns after the transition than the benchmark group. However, studies that address interventions to reduce school transition stress are scarce and were mainly developed in the United Kingdom or the United States (Symonds, 2015), countries where this transition usually takes place after the fifth or sixth grade.

Considering the overwhelming data supporting the need to support students’ transition to middle school, it is surprising that in Portugal, most students who make this transition are not supported by formal programs to prepare for the transition (Coelho, Marchante, & Jimerson, 2017; Pereira & Mendonça, 2005). In fact, the most common type of support provided to students during this transition is a one day visit to the new school (Coelho & Figueira, 2011).

Supporting School Transitions in Portugal

Portugal has one of the highest levels of grade retention among European Union countries (DGEEC, 2015), and after each school transition, there is a very noticeable increase in grade retention and school absenteeism (DGEEC, 2015). Grade retention is the practice of requiring a student who has been in a given grade level for a full school year to remain in the same level for the subsequent school year (Jimerson & Ferguson, 2007); in Portugal, students will be retained if they have three or more retention scores, scores of 1 or 2 on a scale ranging from 1 (poor) to 5 (very good) out of nine subjects by the end of the school year. Jimerson and Ferguson (2007) concluded that retaining a child failed to generate long-term positive effects on either socioemotional and achievement outcomes, because it was unlikely to address the multiple factors influencing the students’ poor achievement or adjustment. In Torres Vedras, the municipality where the present study took place, students who came from the same elementary schools, prior to the intervention, had very high levels of fifth-grade retention (13.7%, 4% above national average; DGEEC, 2015), hence the need for intervention. A preliminary study (Coelho & Romão, 2016) identified that students coming from smaller elementary schools tended to have higher levels of fifth-grade retention.

Positive Transition Program

The “Positive Transition” program is a universal program, applied to whole fourth-grade classes, that aims to promote school adjustment in middle school transition. It was developed in response to solicitations from School Grouping Board for interventions to reduce fifth-grade retention. The program is described in detail in Coelho and Figueira (2011).

Current Study

Preliminary results from the first 2 years of program implementation identified promising results in fifth-grade absenteeism and retention (Coelho, 2015); however, there is a need to identify how the program produced these outcomes. Given that the program focuses on reducing stress in middle school transition, it is crucial to establish whether it is achieving this aim.

This study asked the following research questions:

Research Question 1: Will students from smaller fourth-grade classes have larger increases in all dimensions of stress?

Research Question 2: Will participating in the Positive Transition program reduce transition-related stress levels in all dimensions?

Research Question 3: Will the impact of participation in this program be more pronounced in peer-related stress for girls and in teacher–rules related stress for boys?

Research Question 4: Will students from smaller fourth-grade classes benefit more from participating in the program for all stress dimensions?

Research Question 5: Will participating in the Positive Transition program lead to a reduction in the levels of school absenteeism and to an increase in school success (less grade retentions and higher grade average).

Method

Participants

Student characteristics are described in Table 1. A total of 743 fifth-grade students (54% boys) from 63 classes were evaluated in the middle and end of fourth grade and at the end of fifth grade. Control and intervention groups, defined at the class level, were established in each of the six Portuguese public school groupings from the municipality of Torres Vedras (district of Lisbon) that participated in the study. All elementary schools were rural; three of the middle schools were rural setting, and the other three were urban.

Table 1.

Student Characteristics Across Groups and Years of Implementation (n = 831).

Characteristic	Total (%)	Control group, n = 278 (33.5%)	Intervention group, n = 553 (66.4%)
Gender
Boys	447 (53.8%)	148 (53.2%)	299 (54.1%)
Girls	384 (46.2%)	130 (46.8%)	254 (45.9%)
		χ²(1) = 0.634; p > .05
Previous retention rate	66 (7.9%)	21 (7.6%)	45 (8.1%)
		χ²(1) = 0.367; p > .05
Ethnicity
Portuguese	817 (98.3%)	274 (98.6%)	543 (98.2%)
Brazilian	8 (1.0%)	2 (0.7%)	6 (1.1%)
Eastern European descent	6 (0.7%)	2 (0.7%)	4 (0.7%)
Middle school location
Rural	342 (41.2%)	111 (39.9%)	231 (41.8%)
Urban	489 (58.8%)	167 (60.1%)	322 (58.2%)
		χ²(1) = 2.716; p > .05
Cohort (fourth–fifth grades)
2006/2007–2007/2008	291 (35.0%)	99 (35.6%)	192 (34.7%)
2007/2008–2008/2009	273 (32.9%)	91 (32.7%)	182 (32.9%)
2008/2009–2009/2010	267 (32.1%)	88 (31.7%)	179 (32.4%)
Attrition
Retained in fourth grade	26 (3.5%)	7 (2.5%)	19 (3.4%)
		χ²(1) = 0.986; p > .05
Moved to other schools	62 (8.3%)	20 (7.2%)	42 (7.6%)
		χ²(1) = 0.086; p > .05

Students’ age ranged from 10 to 13 years (M age = 10.62, SD = 0.30) at the end of the intervention (fifth grade), when the measures of school absenteeism and school success were also registered. Intervention and control groups were similar in terms of gender, previous retention rates, middle school location and attrition rates (as seen in Table 1), as well as age, t(1145) = 0.65; p > .05. Data were gathered from three different cohorts, each spanning 2 school years, and there were no initial differences in stress levels per cohort: academic stress, F(2, 740) = 2.49, p > .05; peer-related stress, F(2, 740) = 0.11, p > .05; and teacher/rules stress, F(2, 740) = 0.38, p > .05. All students in each fourth-grade classes were included, ranging from eight to 22 (M = 10.17) per elementary school class. There was attrition (88 students, described in Table 1), due to students who had to repeat fourth grade (26 students; 3.5%) or transitioned into private middle schools or schools outside the municipality (62 students; 8.3%). These students had to be excluded from the sample after taking part in the program during the fourth grade due to the very different conditions they experienced (number of known colleagues that accompany the student, specific space settings and activities of the new schools, etc.). Schools were extremely homogeneous in terms of ethnicity (see Table 1), below the 2.6% national average of students with different ethnic backgrounds (DGEEC, 2015).

Measures

Questionário de Avaliação de Stress Escolar (QASE)

The QASE is the Portuguese adaptation of the School Stress Survey (Wenz-Gross et al., 1997) that allows students to report on the occurrence of disturbing events in school throughout the school year, as well as the level of stress that they attribute to those events. The Portuguese version (Pereira & Mendonça, 2005) is composed by 49 items, evaluating various sources of stress organized into three subscales: Academic Stress, such as “Having low grades” (15 items; α = .88, .89 in the present study); Peer-Related Stress, such as “Being ashamed of my appearance” (14 items, α = .86, .87 in the present study); and Teacher/Rules Stress, such as “Being shamed by a teacher” (20 items; α = .90, .92 in the present study). Pereira and Mendonça (2005) conducted an adaptation and validation of the Survey for Portuguese fourth and fifth graders, reporting a similar three-factor structure, good construct validity (significant correlations between subscales and significant correlations with similar constructs), adequate internal consistency (as reported above), and adequate temporal stability (.77 for the total scale).

In the QASE, each item represents an event of potential stress. First, the student had to decide whether that event occurred during that the previous school year. If the event did not occur, a score of 0 was scored. If the event occurred, the student had to indicate the level of stress associated with said event on a 4-point Likert-type scale ranging from 1 (It happened but I was not concerned about it) to 4 (It happened and I was very concerned about it).

School records

Student school records containing grades and the absences for each subject during a school term (three terms per year) were used for the analysis of school success and absenteeism. School records used were from the third (i.e., final) term, which contain information about transition to sixth grade, achieved if the student had three or fewer retention scores. Fifth-grade school records are quantitative (for subjects like mathematics or history), varying between 1 and 5 (with Grades 1 and 2 corresponding to very poor and poor achievement, respectively, and, therefore, are considered cause for grade retention and Grades 3, 4, and 5 corresponding to satisfactory, good, and very good achievement, respectively).

The indicators used for the current study were grades, summed for the five main areas of study (math, Portuguese, history and geography, English, and natural sciences), and absences, which were measured as the number of classes that the student did not attend (these included disciplinary sanctions).

Procedure

Portuguese schools are organized into groupings, composed by a middle school, several elementary schools, and kindergartens, administrated by a School Board. Every year, the same selection criteria were applied, with each School Grouping Board choosing classes from rural elementary school for participation, randomly assigning them to the control and intervention groups and then informing the program developers. All public school groupings in this municipality participated in the study.

Intervention was the Positive Transition program. The program contains twenty 50-min sessions, which are conducted by an educational psychologist (four in total, with at least 3 years of experience). Fifteen of these sessions took place weekly during the second half of fourth grade (February–June); the remaining five sessions took place during the fifth grade. The program was implemented during school schedule; the same educational psychologist accompanied the groups during the fourth and fifth grades and was present in every session. As detailed in Table 2, during fourth grade, sessions are mainly focused on the cognitive aspects of this transition (e.g., analysis of schedules and evaluation sheets), the emotional aspects of the transition (e.g., discussion about the fears and expectations about fifth grade), as well as behavioral skills training (e.g., during a visit to the middle school, students have to follow a fifth-grade schedule and guide the psychologist throughout the school, thereby modeling the desired behaviors). In fifth grade, a session takes place in the beginning of the school year, two at the end of first term and two at the end of year. In those sessions, children discuss the differences between elementary and middle school environments, analyze the difficulties that they experienced at the transition, and learn coping strategies to deal with their responses to transition stressors.

Table 2.

Summary of Session Contents.

Number	Activity	Duration	Aims	Description
1	Presentation	1 hr	Establish empathy between the psychologist and students	The psychologist explains the program, and four self-presentation activities are implemented.
2	Initial assessment	1 hr	Assess baseline scores	Filling out questionnaires.
3	What changes in fifth grade	1 hr	Reduce academic stress, promote academic self-concept	Information (organized into five themes; friends, school grounds, school rules, teachers, and disciplines) is given to the fourth-grade students about what is going to change in fifth grade.
4, 5	Discussing the fifth grade: fears and hopes	2 hr	Reduce rules/teacher stress, promote emotional self-concept	Fourth-grade students are organized into small groups of four or five, and they have to fill out one sheet per theme addressing two questions: “what do you look forward” and “what are you worried about” in fifth grade. In the second activity (bag of fears), students anonymously and individually are asked to write down their greatest fear about going to the fifth grade. Every fear is then discussed, identifying coping strategies.
6	Getting to know the fifth-grade schedule	1 hr	Reduce rules/teacher stress	Explaining and analysis of a fifth-grade schedule. Presentation of the new subjects. Questions and exercises.
7	The fifth-grade evaluation system	1 hr	Reduce academic stress, promote academic self-concept	Explaining and analysis of a fifth-grade evaluation sheet. Discussion about the new evaluation procedures. Questions and exercises.
8, 9	With a little help from fifth-grade teachers	2 hr	Reduce rules/teacher stress, promote academic self-concept	A fifth-grade class director (a role that does not exist in fourth grade) explains what teachers expect from fifth-grade students in each of the areas discussed previously and then answer questions.
10	With a little help from our fifth-grade friends	1 hr	Reduce peer-related stress, promote social and emotional self-concept	Two fifth-grade students who have previously been students in that school and have been successful in the transition to fifth grade explain their experience of transition and answer questions from the fourth-grade students.
11, 12, 13	Visit to middle school	3 hr	Reduce peer-related and rules/teacher stress, promote academic, emotional self-concept and self-esteem	Fourth-grade students are taken on a tour of the middle school by the school psychologist who is delivering the program, visiting different classrooms and services. A peddy-paper where students (in groups of four or five) have to lead the school psychologist following a fifth-grade schedule and answer questions.
14	Revision	1 hr	Consolidate previous lessons	All the program contents are reviewed, summarized, and questions are answered.
15	Intermediate assessment	1 hr	Assess mid-program scores	Filling out questionnaires.

There were three data collection periods: in the first and last (15th) fourth-grade sessions and in the last fifth-grade (20th) session. In fourth grade, the program started in the middle of the school year (February), and students attended 15 weekly program sessions in their regular classes; in fifth grade, students attended the program sessions in groups composed of their former fourth-grade peers. Program implementation was monitored in weekly team meetings in which each of the psychologists submitted class reports (including a checklist that included students’ attendance and a summary indicating which activities were implemented and what materials were used). A progress report was sent every month to the municipality indicating number of sessions conducted, students present, and implementation fidelity. To assess implementation fidelity, the psychologist completed a checklist, identifying whether the activities were implemented, whether the auxiliary materials were used, and whether the allotted times were followed (e.g., in Session 6, the psychologist were asked whether they created a schedule together with the students, then whether they used the schedule in the auxiliary materials, and finally whether they had implemented the each of the six rounds of questions). Implementation fidelity was high (91% in first cohort; 97% and 93% in the second and third). However, the psychologists reported that in the 15th class fidelity was 36.6%, thus an extra fourth-grade session was conducted to complete with all the activities in the manual.

Data Analysis

The t tests were performed to analyze whether there were initial differences in stress levels between the control and intervention groups. Multilevel linear modeling (MLM) was used (SPSS, mixed models) to evaluate the effects of fourth-grade class size (Research Question 1), of the program on stress levels (Research Questions 2, 3, and 4), and on school success and school absenteeism (Research Question 5). There are several advantages to MLM, compared with other analyses, such as the lack of requirement for complete data across time points (Heck, Thomas, & Tabata, 2013). For the variables with repeated measures (stress), parallel growth models were used. The advantage of this procedure is that it allows for the examination of the extent to which concurrent individual changes in the three dimensions of stress are mutually interrelated, as well as the extent to which changes may be explained by the hypothesized individual (gender) and class-level (fourth-grade class size) factors in the model. In a repeated-measures study design, individual scores are nested within individuals, and these, in turn, are nested within classes. As nested data are more likely to correlate within the group, students’ evaluations from the same class are more likely to be highly correlated. A three-level model was run to account for the 743 individuals being nested within three measurements and that the students were nested in 63 classes.

After several analyses, the best model fit was achieved with first-order autoregressive (AR1) as the covariance type for Level 1 and scaled identity for Levels 2 and 3. Great-mean centering was used for fourth-grade class size, a Level 3 predictor. This was the procedure used to create the unconditional model. For Model 0, a growth rate indicator variable was constructed to manage the parallel growth trajectories in outcome measures. There are several ways to code the growth rate variable; after several analyses, the best fit was achieved by coding the growth according to the months when the assessments were obtained, with the first measurement occasion (baseline) coded as 0, the second measurement occasion as 4 (end of the fourth grade), and 16 for the third measurement (end of fifth grade). Model 1 included Level 2 or individual-level predictors (gender), whereas Model 2 was created to evaluate the effect of class-level predictors (fourth-grade class size and condition), and Model 3 included potential moderating effects from the interactions between time, gender, and condition, and between time, fourth-grade class size, and condition.

Heck et al. (2013) noted that in SPSS MIXED, the reference group for a variable entered as a factor is the last category. As condition has been defined as a factor, the intervention group (= 1) is the reference group and the estimate will be for control group (= 0). IBM SPSS Statistics for Windows, Version 20 (IBM Corp., Armonk, NY, USA), was used.

Results

Preliminary Analysis

The intraclass correlations (ICCs) show that there is sufficient variance be explained for each variable at class level to warrant multilevel analysis, following recommendations by Heck et al. (2013) that ICCs higher than .05 justify inclusion: stress, ICC_class = .052; grade retention, ICC_class = .143; grade average, ICC_class = .201; and missed classes, ICC_class = .182.

The descriptive statistics for types of stress by condition are displayed in Table 3. Comparisons regarding initial levels of stress between control and intervention groups showed no significant differences in any dimension: academic stress, t(741) = 0.13; p > .05; peer-related stress, t(741) = −0.76; p > .05; and teacher/rules stress, t(741) = −0.43; p > .05, so there was no need to impose equality constraints for condition as a predictor.

Table 3.

Types of Stress Descriptive Statistics.

Types of stress	Control group (n = 251)			Intervention group (n = 492)
Types of stress	Fourth-grade middle, M (SD)	Fourth-grade end, M (SD)	Fifth-grade end, M (SD)	Fourth-grade middle, M (SD)	Fourth-grade end, M (SD)	Fifth-grade end, M (SD)
Academic stress	20.34 (9.13)	20.90 (10.11)	24.07 (12.03)	20.24 (10.00)	20.63 (11.74)	22.39 (12.14)
Peer-related stress	14.20 (8.22)	14.27 (9.05)	14.94 (10.66)	14.74 (9.61)	14.62 (10.82)	14.02 (10.47)
Teacher/rules stress	20.22 (12.13)	20.70 (13.82)	23.36 (16.58)	20.65 (13.39)	21.23 (16.58)	19.37 (15.71)

The introduction of gender (a Level 2 predictor) showed that there were differential trajectories for all stress dimensions during the whole period of middle school transition with larger increases for girls in academic (β = 0.09, SE = 0.04; t = 2.51, p = .012) and peer-related stress academic (β = 0.13, SE = 0.04; t = 3.48, p = .001), and for boys in teacher/rules stress (β = −0.11, SE = 0.05; t = −2.35, p = .019), as seen in Table 4, Model 1.

Table 4.

Parallel Growth Curve Models for Types of Stress.

Parameters	Unconditional model	Model 0Level 1: Time	Model 1Level 2: Individual	Model 2Level 3: Class
Estimates of fixed effects
Intercept academic stress	21.77 (0.50)***	20.31 (0.55)***	20.31 (0.55)***	20.20 (0.55)***
Intercept peer-related stress	13.75 (0.49)***	14.61 (0.55)***	14.61 (0.55)***	14.57 (0.55)***
Intercept teacher–rule stress	20.22 (0.50)***	20.66 (0.55)***	20.66 (0.55)***	20.68 (0.55)***
Academic Stress × Time	—	0.18 (0.03)***	0.14 (0.03)***	0.25 (0.08)*
Peer-Related Stress × Time	—	−0.01 (0.03)	−0.07 (0.03)*	0.07 (0.08)
Teacher–Rule Stress × Time	—	−0.01 (0.03)	0.04 (0.03)	0.16 (0.09)
Academic Stress × Time × Gender	—	—	0.09 (0.04)*	0.10 (0.04)**
Peer-Related Stress × Time × Gender	—	—	0.13 (0.04)**	0.13 (0.04)***
Teacher–Rule Stress × Time × Gender	—	—	−0.11 (0.05)*	−0.09 (0.05)
Academic Stress × Time × Condition	—	—	—	0.12 (0.05)**
Peer-Related Stress × Time × Condition	—	—	—	0.10 (0.05)
Teacher–Rule Stress × Time × Condition	—	—	—	0.31 (0.05)***
Academic Stress × Time × Class Size	—	—	—	−0.02 (0.01)*
Peer-Related Stress × Time × Class Size	—	—	—	−0.02 (0.01)*
Teacher–Rule Stress × Time × Class Size	—	—	—	−0.03 (0.01)**
Estimates of covariance parameters
Repeated measures	147.80 (4.29)***	147.02 (4.27)***	146.58 (4.26)***	145.92 (4.25)***
AR1 ρ	0.71 (0.01)***	0.71 (0.01)***	0.71 (0.01)***	0.71 (0.01)***
Intercept fourth-grade class	8.12 (1.59)***	7.78 (1.60)***	7.69 (1.61)***	7.07 (1.60)***
Deviance (−2_{loglikelihood})	48,437.991	48,388.586	48,364.692	48,289.191
Δ_−2LL		49.41***	23.89***	75.47***
Number of estimated parameters	6	9	12	18

Note. AR1 = first-order autoregressive; Estimate (Standard Error); Δ_−2LL = difference between tested model and previous model with corresponding dfs.

p < .05. **p < .01. ***p < .001.

Fourth-Grade Class Size Effects on Stress Levels

The introduction of Level 3 predictors (condition and fourth-grade class size) led to a decrease of 8.1% in Level 3 (class) variance in the model. To test Research Question 1, Model 2 (see Table 4) included the interaction between fourth-grade class size and time for all three types of stress and, after adjusting for gender and condition, fourth-grade class size still had an effect on all types of stress with students from larger fourth-grade classes having a lower increase in all dimensions of stress during middle school transition: academic stress, β = −0.01, SE = 0.005, t = −2.08, p = .038; peer-related stress, β = −0.01, SE = 0.005, t = −2.37, p = .009; and teacher/rules stress, β = −0.02, SE = 0.006, t = −2.91, p = .004.

Program Effects on Stress Levels

To test Research Question 2, Model 2 (see Table 4) included the interaction between condition and time for all three types of stress and, after adjusting for gender and fourth-grade class size, condition still had a significant effect upon positive effect on academic stress (β = 0.12, SE = 0.05, t = 2.62, p = .009) and particularly on teacher/rules stress (β = 0.31, SE = 0.05, t = 6.09, p < .001).

Program effects by gender

In Model 2, after adjusting the model for Level 3 predictors (condition and fourth-grade class size), the interaction between time and gender for teacher/rules stress was no longer significant (see Table 4, Model 2). In addition, to test Research Question 3, we investigated possible interaction effects between Level 2 and Level 3 predictors (gender and condition) during the full-time of the intervention (Model 3). No significant interactions between time, condition, and gender were found for any stress dimension: academic stress (β = −0.08, SE = 0.08, t = −1.03, p > .05), peer-related stress (β = 0.10, SE = 0.08, t = 1.38, p > .05), and teacher/rules stress (β = 0.01, SE = 0.10, t = 0.09, p > .05). In addition, the fit of the model did not improve significantly when the interactions between condition and gender and condition and fourth-grade class size were added, Δχ²(6) = 5.91, p > .05.

Program effects by fourth-grade class size

To test Research Question 4, Model 3 also analyzed whether there were any significant interaction between Level 3 predictors (condition and fourth-grade class size) during the full-time of the intervention. No significant interactions between time, condition, and fourth-grade class size were found for any stress dimension: academic stress (β = 0.02, SE = 0.01, t = 1.57, p > .05), peer-related stress (β = 0.01, SE = 0.01, t = 0.44, p > .05), and teacher/rules stress (β = 0.002, SE = 0.01, t = 0.18, p > .05).

Program Effects on School Success and School Absenteeism

Descriptive statistics for indicators of school success and school absenteeism are displayed in Table 5.

Table 5.

Means and Standard Deviation for School Success and School Absenteeism Indicators by Condition and Gender.

	Control group			Intervention group
	Boys (n = 136)	Girls (n = 115)	Total (n = 251)	Boys (n = 267)	Girls (n = 225)	Total (n = 492)
Success Indicator	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Transition to sixth grade	83.9%	93.1%	87.6%	92.9%	97.8%	95.3%
Grade average	15.54 (3.60)	16.96 (3.39)	16.12 (3.58)	16.64 (3.36)	17.91 (2.95)	17.22 (3.24)
Total absences	13.57 (22.14)	10.68 (17.43)	12.10 (20.37)	6.44 (15.63)	3.20 (6.09)	4.96 (12.32)

The addition of Level 2 predictor (gender) increased model fit for all variables: transition to sixth grade, Δ_−2LL(1) = 82.04, p < .001; grade average, Δ_−2LL(1) = 217.67, p < .001; and missed classes, Δ_−2LL(1) = 49.54, p < .001 (Model 1, see Tables 6, 7, and 8). Significant results were found for all three variables indicating that girls had higher grade promotion, higher grade averages, and less missed classes than boys. To test Research Question 5, Level 3 predictors were added (condition and fourth-grade class size), and this led to better fitting models for all variables (Model 2, see Tables 6, 7 and 8), as well as a greater decrease in Level 3 variance, 60% for transition to sixth grade, 24% for grade average, and 36.2% for missed classes.

Table 6.

Multilevel Model Analysis Models for Transition to Sixth Grade.

Parameters	Model 0	Model 1	Model 2	Model 3
Parameters	Null	Level 1: Individual	Level 2: Class	Cross-level interactions
Estimates of fixed effects
Intercept	0.92 (0.01)***	0.90 (0.01)***	0.91 (0.03)***	0.92 (0.03)***
Gender (girls = 1)	—	0.06 (0.01)***	0.06 (0.01)***	0.05 (0.01)***
Condition (control = 0)	—	—	−0.08 (0.02)***	−0.10 (0.02)***
Fourth-grade class dimension	—	—	0.002 (0.002)	0.001 (0.002)
Gender × Condition	—	—	—	0.03 (0.01)*
Estimates of covariance parameters
Residual	0.06 (0.001)***	0.06 (0.001)***	0.06 (0.001)***	0.06 (0.001)***
Intercept fourth-grade class	0.01 (0.001)***	0.01 (0.001)***	0.004 (0.001)***	0.004 (0.001)***
Deviance (−2_{loglikelihood})	528.462	446.420	427.137	421.991
Δ_−2LL		82.04***	19.28***	5.15*
Number of estimated parameters	3	4	6	7

Note. Estimate (Standard Error); Δ_−2LL = difference between tested model and previous model with corresponding dfs.

p < .05. **p < .01. ***p < .001.

Table 7.

Multilevel Model Analysis Models for the Grade Average Variable.

Parameters	Model 0	Model 1	Model 2	Model 3
Parameters	Null	Level 1: Individual	Level 2: Class	Cross-level interactions
Estimates of fixed effects
Intercept	16.72 (0.20)***	16.22 (0.19)***	15.80 (0.58)***	15.85 (0.58)***
Gender (girls = 1)	—	1.15 (0.08)***	1.15 (0.08)***	1.06 (0.09)***
Condition (control = 0)	—	—	−1.30 (0.35)***	−1.40 (0.36)***
Fourth-grade class dimension	—	—	0.08 (0.04)	0.07 (0.04)
Gender × Condition	—	—	—	0.25 (0.16)
Estimates of covariance parameters
Residual	9.33 (0.16)***	9.03 (0.16)***	9.03 (0.16)***	9.03 (0.16)***
Intercept fourth-grade class	2.35 (0.44)***	2.16 (0.40)***	1.64 (0.31)***	1.64 (0.31)***
Deviance (−2_{loglikelihood})	34,113.283	33,895.617	33,878.549	33,876.239
Δ_−2LL		217.67***	17.07***	2.31
Number of estimated parameters	3	4	6	7

Note. Estimate (Standard Error); Δ_−2LL = difference between tested model and previous model with corresponding dfs.

p < .05. **p < .01. ***p < .001.

Table 8.

Multilevel Model Analysis Models for the Missed Class Variable.

Parameters	Model 0	Model 1	Model 2	Model 3
Parameters	Null	Level 1: Individual	Level 2: Class	Cross-level interactions
Estimates of fixed effects
Intercept	7.86 (0.89)***	9.01 (0.89)***	7.80 (2.46)**	8.16 (2.47)**
Gender (girls = 1)	—	−2.63 (0.37)***	−2.62 (0.37)***	−3.21 (0.46)***
Condition (control = 0)	—	—	8.16 (1.49)***	7.42 (1.53)***
Fourth-grade class dimension	—	—	−0.15 (0.19)	−0.15 (0.18)
Gender × Condition	—	—	—	1.77 (0.79)*
Estimates of covariance parameters
Residual	212.78 (3.70)***	211.26 (0.16)***	211.28 (0.16)***	211.09 (0.16)***
Intercept fourth-grade class	47.40 (8.93)***	45.57 (8.61)***	29.06 (5.63)***	29.40 (5.70)***
Deviance (−2_{loglikelihood})	55,019.675	54,970.133	54,943.457	54,938.458
Δ_−2LL		49.54***	26.68***	5.00*
Number of estimated parameters	3	4	6	7

Note. Estimate (Standard Error); Δ_−2LL = difference between tested model and previous model with corresponding dfs.

p < .05. **p < .01. ***p < .001.

In the final models, after adjusting for gender, condition, fourth-grade class size, and the interaction between gender and condition, we concluded that students from the intervention group had fewer grade retentions (β = −0.10, SE = 0.02, t = −5.04, p < .001) and higher average grades (β = −1.40, SE = 0.36, t = −3.93, p < .001) than the control group students (Model 3, see Tables 6 and 7). Intervention group students also missed significantly fewer classes than control group students (β = 7.42, SE = 1.53, t = 4.84, p < .001; see Table 8, Model 3). There was no significant impact of the fourth-grade class size (Level 3 predictor) on any outcome.

Program effects on school success and school absenteeism, by gender

There were two significant cross-level interactions between gender and condition, indicating that the boys benefited more from the program in the reduction of grade retention (β = 0.03, SE = 0.01, t = 2.27, p = .027; see Table 6, Model 3) and missed classes (β = 1.77, SE = 0.79; t = 2.24, p = .025; see Table 8, Model 3).

Discussion

The current study examined the effectiveness of a school adjustment program for middle school transition, with the aim of preventing the increase in stress levels, grade retention, and school absenteeism. Results showed that participation in the Positive Transition program led to a positive impact on two dimensions of stress (academic and teacher/rules), and to fewer grade retentions, class absences, and higher grade averages than students in the control group, thus supporting the second (partially) and fifth research questions. A possible explanation for these results can be found in Qualter et al. (2007), who reported that the use of more appropriate coping strategies probably motivates students to work harder, leading to better academic achievement. The improvement in academic and teacher/rules stress probably led to fewer absences in the intervention group. The positive results found in the stress levels are in line with Bloyce and Frederickson (2012), who reported positive benefits related to transition concerns from a middle school transition intervention. They are also particularly important because transition in Portugal takes place at an earlier age than in most other countries (e.g., the United Kingdom or the United States).

Regarding fourth-grade school size, the results supported Research Question 1; students from smaller fourth-grade classes displayed higher increases in all dimensions of stress, even after adjusting for gender and condition. The results are aligned with Cotterell (1992), who concluded that children were less anxious when transition occurred between similarly sized school, and with Symonds (2015), who suggested that having more friends making the same transition made children feel less vulnerable in middle school. Results highlighted the benefits of transitioning into a new school with a larger cohort of friends, which likely reduces the likelihood of disrupting existing friendship networks and provides a safer base for establishing new friendships. However, students from smaller fourth-grade classes did not have more benefits from participating in the program; therefore, results did not support the fourth research question.

The results also showed several gender differences in middle school transitions. Girls displayed higher rates of transition to the sixth grade, higher grade average, and fewer class absences than boys. However, after adjusting for condition and fourth-grade class size, girls also reported a greater increase in academic and peer-related stress during the transition. Girls displaying more peer-related stress in middle school transition are commonly found in the literature (Coelho & Romão, 2016), with Chung et al. (1998) suggesting that, for girls, the stressfulness of the transition is expressed via internalization, rather than externalized in behavioral problems. However, no gender differences were found in the stress levels during middle school transition after adjusting for condition, gender, and fourth-grade class size, thus negating the third research question, although boys benefited more from participation in the program with larger increases in the transition rates to sixth grade and greater decreases in missed classes. The benefits boys showed (relative to girls) from participation in the program in transition to sixth grade probably stemmed from a ceiling effect, given that girls already had a very high rate of transition to sixth grade, even without participating in the program.

The results demonstrated that the Positive Transition program helped to promote school success in fifth grade and mitigated the decline in school performance that several authors (Coelho & Romão, 2016; Goldstein et al., 2015; Ryan et al., 2013) reported and which was noticeable in the control groups. Therefore, the program was effective in creating a more positive adaptation to the new school context, both emotionally and academically. These results highlight the importance of intervening because, as several authors concluded (Akos & Galassi, 2004b; Pietarinen, 2000; Qualter et al., 2007; Ryan et al., 2013), transition to middle school is difficult for students, particularly in Portugal where early transition is a feature of the school context (Coelho & Romão, 2016; Pereira & Mendonça, 2005). Without intervention, as observed in the control group, middle school transition led to increased stress levels, lower attendance rates, and lower grades. The intervention helped to set in motion the conditions for future school success because, according to Jimerson and Ferguson (2007), children who are at risk because of poor achievement or adjustment require additional resources or services to facilitate achievement trajectories.

Limitations

It would be important to complement the results of this study with an in-depth analysis about the specific concerns (e.g., going to a bigger school; making new friends) that students reported in fourth grade and how they relate to levels of stress pretransition. Another limitation was that implementation fidelity was only assessed through self-reports, which limits its validity and may be subject to inaccuracy or bias. As there was no quantitative grading established in Portuguese elementary school, it also was not possible to control for previous academic achievement level. In addition, the control group, even though sizable, was much smaller than the intervention group.

Future Directions

The current study is an initial step for understanding the impact of transition with a younger population. Future studies should involve comparisons with students from other countries making this transition at a similar age point (e.g., Germany, K-4 schools in the United States), to analyze whether the conclusions about students from smaller classes having a greater increase in stress dimensions are generalizable.

In Portugal, it is well established that the risk of retention increases with previous grade retentions (Coelho & Figueira, 2011). As such, if it is possible to delay retention, it may also be possible to prevent repeated grade retentions. Therefore, future studies should follow up with the groups to assess whether the effects of the intervention on stress levels lasted and whether this contributed to lower grade retention in middle school. Future studies should also examine the role of initial achievement levels on later performance. Also, it would be important to analyze whether students who are supported during the transition to middle school are more likely to experience improved self-esteem and social and academic self-concept, as predicted by previous research (Arens et al., 2013).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Municipality of Torres Vedras.

ORCID iD

Vítor Alexandre Belo de Moura Pereira Coelho

References

Akos

Galassi

J. P.

(2004a). Gender and race as variables in psychosocial adjustment to middle and high school. The Journal of Educational Research, 98, 102–108. doi:10.3200/JOER.98.2.102-108

Akos

Galassi

J. P.

(2004b). Middle and high school transitions as viewed by students, parents and teachers. Professional School Counseling, 7, 212–222.

Anderson

Jacobs

Schramm

Splittberger

(2000). School transitions: Beginning of the end or a new beginning? International Journal of Educational Research, 33, 325–339. doi:10.1016/S0883-0355(00)00020-3

Arens

A. K.

Yeung

A. S.

Craven

R. G.

Watermann

Hasselhorn

(2013). Does the timing of transition matter? Comparison of German students’ self-perceptions before and after transition to secondary school. International Journal of Educational Research, 57, 1–11. doi:10.1016/j.ijer.2012.11.001

Bloyce

Frederickson

(2012). Intervening to improve the transfer to secondary school. Educational Psychology in Practice, 28, 1–18. doi:10.1080/02667363.2011.639345

Chung

Elias

Schneider

(1998). Patterns of individual adjustment changes during middle school transition. Journal of School Psychology, 36, 83–101. doi:10.1016/S0022-4405(97)00051-4

Coelho

V. A.

(2015). Promoção do sucesso e ajustamento escolar: Estudos sobre a eficácia e efectividade de programas de desenvolvimento socioemocional para alunos do 4º ao 9º ano. [Enhancing school success and school adjustment: Studies analyzing the efficacy and effectiveness of social and emotional learning programs for students from the 4th to the 9th grade]. (Unpublished doctoral dissertation). Coimbra: University of Coimbra.

Coelho

V. A.

Figueira

A. P.

(2011). Project “Positive Attitude”: Promoting school success through social and emotional abilities development. Design for elementary and middle school students, in Portugal. Interamerican Journal of Psychology, 45, 185–192.

Coelho

V. A.

Marchante

Jimerson

S. R.

(2017). Promoting a positive middle school transition: A randomized-controlled treatment study examining self-concept and self-esteem. Journal of Youth and Adolescence, 46, 558-569. doi:10.1007/s10964-016-0510-6

10.

Coelho

V. A.

Romão

A. R.

(2016). Stress in Portuguese middle school transition: A multilevel analysis. The Spanish Journal of Psychology, 19. doi:10.1017/sjp.2016.61

11.

Cotterell

J. L.

(1992). School size as a factor in adolescents’ adjustment to the transition to secondary school. Journal of Early Adolescence, 12, 28–45. doi:10.1177/0272431692012001002

12.

Direção Geral de Estatísticas da Educação e da Ciência. (2015). Educação em números 2014 [Education in numbers 2014]. Lisboa, Portugal: Author.

13.

Eccles

J. S.

Roeser

R. W.

(2009). Schools, academic motivation, and stage-environment fit. In Lerner

R. M.

Steinberg

L. D.

(Eds.), Handbook of adolescent psychology (3rd ed., pp. 404–434). Hoboken, NJ: John Wiley. doi:10.1002/9780470479193.adlpsy001013

14.

Goldstein

S. E.

Boxer

Rudolph

(2015). Middle school transition stress: Links with academic performance, motivation and school experiences. Contemporary School Psychology, 19, 21–29. doi:10.1007/s40688-014-0044-4

15.

Heck

R. H.

Thomas

S. L.

Tabata

L. N.

(2013). Multilevel and longitudinal modeling with IBM SPSS (2nd ed.). London, England: Routledge.

16.

Jimerson

S. R.

Ferguson

(2007). A longitudinal study of grade retention: Academic and behavioral outcomes of retained students through adolescence. School Psychology Quarterly, 22, 314–339. doi:10.1037/1045-3830.22.3.314

17.

Lane

K. L.

Oakes

W. P.

Carter

E. W.

Messenger

(2015). Examining behavioral risk and academic performance for students transitioning from elementary to middle school. Journal of Positive Behavior Interventions, 17, 39–49. doi:10.1177/1098300714524825

18.

Lazarus

R. S.

(2000). Evolution of a model of stress, coping, and discrete emotions. In Rice

V. H.

(Ed.), Handbook of stress, coping and health: Implications for nursing research, theory and practice (pp. 195–222). Thousand Oaks, CA: SAGE.

19.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal and coping. New York, NY: Springer.

20.

Pereira

Mendonça

(2005). O Stresse escolar na transição de escolas do 1º para o 2º ciclo do ensino básico: a versão portuguesa do questionário de avaliação do stress escolar [School stress in the transition to middle school: The Portuguese version of the School Stress Survey]. Psicologia, Educação e Cultura, 9, 89–107.

21.

Pietarinen

(2000). Transfer to and study at secondary school in finish school culture: Developing schools on the basis of pupils’ experiences. International Journal of Educational Research, 33, 383–400. doi:10.1016/S0883-0355(00)00024-0

22.

Qualter

Whiteley

H. E.

Hutchinson

J. M.

Pope

D. J.

(2007). Supporting the development of emotional intelligence competencies to ease the transition from primary to high school. Educational Psychology in Practice, 23, 79–95. doi:10.1080/02667360601154584

23.

Rice

Frederickson

Seymour

(2011). Assessing pupil concerns about transition to secondary school. British Journal of Educational Psychology, 81, 244–263. doi:10.1348/000709910X519333

24.

Ryan

A. M.

Shim

S. S.

Makara

K. A.

(2013). Changes in academic adjustment and relational self-worth across the transition to middle school. Journal of Youth and Adolescence, 42, 1372–1384. doi:1007/s10964-013-9984-7

25.

Symonds

(2015). Understanding school transition: What happens to children and how to help them. London, England: Routledge.

26.

Symonds

Galton

(2014). Moving to the next school at age 10-14 years: An international review of psychological development at school transition. Review of Education, 2, 1–27. doi:10.1002/rev3.3021

27.

Wenz-Gross

Siperstein

G. N.

Untch

A. S.

Widaman

K. F.

(1997). Stress, social support, and adjustment of adolescents in middle school. Journal of Early Adolescence, 17, 129–151. doi:10.1177/0272431697017002002

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

0.08 MB