Observational Analysis of the Organization of On-Task Behavior in the Classroom Using Complementary Data Analysis

Análisis observacional de la organización del comportamiento en la tarea en el aula utilizando complementariedad de análisis de datos

 

 

Carlos Santoyo¹, Gudberg K. Jonsson², M. Teresa Anguera³ and José Antonio López-López⁴

¹ Universidad Nacional Autónoma de México. Ciudad de México (México).
² Iceland University. Rejkyavík (Iceland).
³ Universidad de Barcelona. Barcelona (Spain).
⁴ Universidad de Málaga. Málaga (Spain).

We acknowledge the support of the Mexican Science and Technology Board CONACYT (project 178383), PAPIIT/UNAM (project IN306715) and thank L. Colmenares, Y. Torres, and N. Xicotencatl for their help.

We gratefully acknowledge the support of the Spanish government (Ministerio de Economía y Competitvidad) within the Projects Avances metodológicos y tecnológicos en el estudio observacional del comportamiento deportivo [Grant PSI2015-71947-REDT; MINECO/FEDER, UE] (2015-2017), and La actividad física y el deporte como potenciadores del estilo de vida saludable: evaluación del comportamiento deportivo desde metodologías no intrusivas [Grant DEP2015-66069-P; MINECO/FEDER, UE] (2016-2018).

We gratefully acknowledge the support of the Generalitat de Catalunya Research Group (GRUP DE RECERCA E INNOVACIÓ EN DISSENYS [GRID]). Tecnología i aplicació multimedia i digital als dissenys observacionals], [Grant 2014 SGR 971].

Correspondence

 

 

---

ABSTRACT

The aim of this study was to analyze the organization of on-task behavior in the classroom. Four observational methodology techniques- T-pattern detection, lag sequential analysis, trend analysis, and polar coordinate analysis-were used to study the organization of on-task and offtask behavioral patterns during class time in a primary school setting. The specific objective was to detect and explore relationships between on-task behavior and different social interaction categories in relation to the actual distribution of activities in a real-life classroom setting. The study was conducted using the behavioral observation system for social interaction SOC-IS and the software programs Theme (version 6, Edu), SDIS-GSEQ (version 4.1.2), HOISAN (version 1.6), and STATGRAPHICS (version 6). We describe the results obtained for the four techniques and discuss the methodological implications of combining complementary techniques in a single study.

Key words: Observational methodology; social interaction; academic engagement; T-pattern; lag sequential analysis; trend analysis; polar coordinates.

---

RESUMEN

El objetivo de este estudio es analizar la organización de la actividad académica en el aula de clase. Cuatro técnicas de análisis de datos utilizadas en metodología observacional -detección de T-Patterns, análisis secuencial de retardos, análisis de tendencias, y análisis de coordenadas polares- han permitido estudiar como los escolares de Primaria distribuyen sus actividades en el aula. De forma específica, se pretendía detectar y explorar las relaciones entre las conductas relativas al trabajo académico y diferentes categorías de interacción social respecto al uso del tiempo en el contexto de la vida cotidiana en el aula. El estudio se llevó a cabo mediante el instrumento de observación SOC-IS, focalizado en la interacción social, y se utilizaron los programas informáticos THEME (versión 6, Edu), SDIS-GSEQ (versión 4.1.2), HOISAN (versión 1.6), y STATGRAPHICS (versión 16). Se describen los T-Patterns, patrones de conducta, tendencias y vectores obtenidos, sí como las implicaciones metodológicas de la estrategia propuesta.

Palabras clave: Metodología observacional; interacción social; persistencia académica; detección de T-Patterns; análisis secuencial de retardos; análisis de tendencias; análisis de coordenadas polares.

---

 

Introduction

In this paper, we analyze the organization of behavioral patterns of children in a primary school classroom by identifying behaviors that favor or interfere with on-task activity during academic instruction.

The Teaching and Learning International Survey (TALIS) is a cross-country survey conducted by the Organization for Economic Co-operation and Development (OECD) to investigate how, among other things, teachers distribute their time in the classroom, which is, logically, related to time spent by children on academic activities (TALIS, 2013). The survey showed that one of four teachers spend at least 30% of their time dealing with disruptive behavior, interruptions, and administrative tasks. In Mexico, teachers reported that they spent just 69% of their classroom time on learning activities (vs 80% in other OECD countries). Secondary school teachers, in turn, reported spending 345 hours a year resolving student conflicts, imposing order, and doing paperwork, which is a considerable amount of time that could be spent on instruction. Findings such as these are cause for concern and of particular relevance in several OECD countries, especially in terms of the implications for public policies in Latin America (Martinic, 2015). While information of this type seems crucial for education planning and evaluation purposes, research in this area to date has focused mainly on sources of indirect data rather than on the use of direct observation to investigate realtime activity patterns of teachers, and equally importantly, children in the classroom.

The present study analyzes streams of behavior exhibited by children performing academic tasks in the classroom. We employed an observational methodology design to gather information on the organization of behavioral patterns in a primary school classroom. Observational methodology has emerged in recent years as a highly effective and useful technique for analyzing behaviors and events in a wide range of fields. Adequate use of this scientific method provides a rigorous analytical framework for addressing specific research questions using one or more research designs inherent to observational methodology.

Most observational methodology studies to date have analyzed data from a single perspective, i.e., using a single method, although there is growing interest in the combined use of complementary methods to unlock the informative and predictive power hidden within large volumes of data collected using direct observation methods (Tarragó, Iglesias, Lapresa, & Anguera, 2016; Tarragó, Iglesias, Lapresa, Anguera, Ruiz-Sanchis, & Arana, in press). Such approaches can only contribute to enhancing substantive and methodological richness, and consequently improving our understanding and interpretation of empirical evidence (data). Although each method has its own characteristics and purpose, the "cross-fertilization" of methods is likely to provide additional and complementary insights into given research questions. Furthermore, the combination of distinct technical and methodological approaches provides a broad perspective from which to contemplate different angles and aspects of the same question.

For the purpose of the present study, it was necessary to identify and analyze the dynamics of behaviors and behavioral patterns within and related to on-task academic activities. In situ observation of on-task behavior as an indicator of classroom motivation and social ecology (Heward, 1994) or of frequency of activity interruption (Santoyo, 2010) is an adequate strategy for this purpose. Previous studies using a similar approach have shown the high level of flexibility offered by observational methodology in the classroom (Santoyo & Anguera, 1992; Otero & Haut, 2016; Razo, 2015; Santoyo, Fabián, & Espinosa, 2000).

Earlier studies have analyzed control exerted by teachers over time spent on academic tasks by students and the effects of on-task behavior and contingent teacher attention on academic engagement (Abramowitz, O'Leary & Futter-sak, 1988; Berk & Landau, 1993; Heller & White, 1975). The effects of reinforcement strategies on academic activity have also been analyzed (Martens, 1990; Martens, Halperin, Rummel, & Kilpatrick, 1990; Santoyo et al, 2000). Forms and means of teacher-student interactions can be described by the matching law (Torres, 2012; Delgado, 2013), and two very recent studies have shown that reinforcement interventions focused on self-regulation can improve on-task behavior (Norris, 2016; Slattery, Crosland, & Iovannone, 2016). However, it is also necessary to collect information on classroom events that occur that influence the development of diverse behavioral patterns, and in particular the behavior stream of individual students and their social interaction with peers.

One way of studying behavioral patterns in the classroom is to analyze the frequency of transition from one behavior to the next. While many children typically "jump" from one activity to another, others are relatively persistent when it comes to task completion. Previous studies of children at primary schools, for example, have shown that most children change activities about three or four times every minute, and that on average they dedicate just 25% to 30% of available class time to academic activities. In other words, around three-quarters of instruction time is spent on nonacademic and social activities (Santoyo et al., 2000; Santoyo, Morales, Colmenares, & Figueroa, 2007). Similar results have been reported for other classroom settings (Razo, 2015). It is therefore important not only to determine what favors on-task behavior and how long children spend on task, but also to assess behavioral patterns over a period of time.

To address these questions, we built on former research involving archival data (Elder, Pavalko, & Clipp, 1993) and defined new research questions based on information compiled from the databases of the Longitudinal Coyoacán Study (Santoyo, 2007; Santoyo, Espinosa, & Bachá, 1994, 1996; Santoyo & Colmenares, 2012). The availability of consistent, robust, and reliable information permits the formulation of new questions that can be answered with adequately collected data.

The aim of this study was to use four complementary observational methodology techniques to analyze relationships between episodes of on-task behavior and different social interaction categories and to transfer our results to the classroom in order to better understand the development of student behavioral patterns.

 

Method

Design

We undertook an observational methodology study, which is appropriate for analyzing spontaneous behavior in a natural setting. The specific design used was an I/F/M design (Anguera, Blanco-Villaseñor, & Losada, 2001), where I refers to idiographic (analysis of data for 28 students although in this study we focus on just one student), F refers to follow-up (data collected over eight sessions), and M refers to multidimensional (assessment of multiple levels of response). The multidimensional nature of the study influenced subsequent decisions regarding the content of the observation instrument (Anguera, Magnusson, & Jonsson, 2007) and the nature of the data (Bakeman, 1978).

Participants

We contemplated both molar and molecular views of the behaviors analyzed. From the molar perspective, we briefly described the overall use of classroom time by 28 first-year primary school students (18 boys and 10 girls) in a public school in Mexico City, Mexico. From the molecular perspective we focused on the individual behaviors of one student in particular (the target student, coded as fdo1f) as a case study. Details of the sampling method and other aspects of the Longitudinal Coyoacan Study are described elsewhere (Santoyo & Espinosa, 2006; Santoyo, 2007). The research project was approved by an ethics committee at the university of the first author and the study was approved by the school authorities. Informed consent was obtained from all the children's parents.

Instruments

1. Observation instrument. The observation instrument used to analyze and code the behaviors of the children was the behavioral observation system for social interaction, known as SOC-IS, according to its acronym in Spanish (Santoyo et al, 1994). The instrument is formed by five dimensions or criteria (Classmate, Initiation, Episode, Task, Area), each of which is broken down into a system of exhaustive, mutually exclusive categories (Table 1). SOC-IS is a valid, reliable, flexible, and viable instrument with a long track record and guarantees of quality, including between-observer agreement rates of over 0.8 (Bakeman & Gottman, 1986) and generalizability rates of over 0.9 for trained observers, individuals (sample representativeness), and number of sessions (Espinosa, Blanco-Villaseñor, & Santoyo, 2006).

Characteristics of SOC-IS:

a) Numerous exhaustive and mutually exclusive behavioral categories, with the following response criteria: Episode, Initiation, and Area

b) Representative category system based on actions performed by children in the classroom (academic tasks, social interaction, group play, individual play, nonacademic behavior)

c) Recording of event-based data based on events that occur within predefined intervals

d) Detection of direction of social interaction to identify initiator of action

2. Recording instrument:

The freely available software program SDIS-GSEQ version 4.1.2. (Bakeman & Quera, 1996, 2011) [http://www.ub.es/comporta/sg/sg_s_download.htm], loaded with SOC-IS, was used to analyze and code the video footage of the sessions via time-based data.

3. Data analysis instruments. To meet the study objective, which required the use of four analytical techniques, we used the following four software programs:

a. THEME 6 Edu (Magnuson, 1996, 2000, 2005) for the T-Pattern analysis (free software program). This software is free. Table 2 shows a screenshot of the data loaded into THEME.

b. SDIS-GSEQ version 4.1.2. (Bakeman & Quera, 1996, 2011) [http://www.ub.es/comporta/sg/sg_s_download.htm] to calculate the adjusted residuals needed to analyze sequential behaviors by lag sequential analysis (free software program)

c. HOISAN version 1.6 (Hernández-Mendo, López-López, Castellano, Morales-Sánchez, & Pastrana, 2012; Hernández-Mendo et al., 2014) to calculate polar coordinates according to the original proposal of Sackett (1980) and the concept of genuine retrospectivity proposed by Anguera (1997). This software program is also free and displays results in the form of polar coordinate maps. Table 3 shows a screenshot of the observation instrument SOC-IS loaded into HOISAN.

d. STATGRAPHICS version 16 to generate linear regression equations and display the corresponding trends in graph format.

 

 

 

Procedure

Eight 15-minute classroom sessions were analyzed and coded in situ by trained observers in SDIS-GSEQ using timed data (type IV data according to the data types described by Bakeman [1978]) derived from within-session observational sampling using 5-second intervals.

Datasets with interobserver agreement levels of less than 80% were eliminated and a new dataset was generated. Interobserver agreement (kappa statistic) for the behaviors of the target child was 0.94 (Cohen, 1960, 1968). This analysis was performed on the same day as the field session, following coding of the data.

 

Results

In this section we investigate the convergence between the four data analysis techniques used to investigate the relationships between on-task behavior (eac category in SOC-IS) and the other social interaction categories in the instrument. We chose the eac category to represent on-task behavior as a key indicator of academic motivation and the impact of teacher-led activities. In studies of the use of class time, it is obviously important to analyze how much time children spend on instructional practices. Our study differs from previously cited studies in that it provides information on specific interactions that are generally broadly classified as off-task activities, thereby providing insights into how a specific child and his peers contribute to the persistence of academic engagement in the classroom.

At the molar level, the 28 children switched behaviors on average three or four times a minute and 25% of the transitions were in response to an interaction initiated by another child. The children generally spent less than 30 seconds on task at a time, and the likelihood of switching from an on-task to an off-task activity was linearly dependent on the time spent on the first task. In other words, the longer the time spent on eac, the less likely the child was to abandon the task and vice versa. For more information on the molar distribution of behavioral transitions we recommend consulting Santoyo (2006).

Although we analyzed diachronic relationships between eac and the other categories for the 28 children over eight sessions, in this next section, we focus on the results obtained for the target student fdo1f. The boy spent just 19% of the total class time recorded on eac, the rest of his time was divided between social interaction (43%) and other activities (38%). The specific behavioral patterns detected using each of the four techniques are described below.

T-Pattern detection

The first analysis was T-pattern detection, which is used to identify hidden patterns within sequential datasets (Magnusson, 1996, 2000, 2005). To meet the requirements of the technique and the software program (Theme v. 6 Edu), we prepared the vvt.vvt file corresponding to the observation instrument and the respective .txt data files, arranged in a single block. A screenshot from the program is shown in Figure 1.

 

 

For the first analysis we search for patterns occurring in at least 50 % of the observational files, using a significance level of p < .05. The pattern detection resulted in 133 different T-patterns. To narrow down the selection using all the calculations for the target student, we prioritized T-patterns first according to their level of statistical significance and second according to their relationship with the aims of the study (i.e. T-patterns related to the performance of academic tasks in the classroom, i.e., the connection between eac and ml). The category m1 refers to the place assigned to the student in the classroom, i.e., his desk. Table 3 shows all the areas considered for the general study, but for the purpose of this analysis, we used m1 to investigate whether the target student interacted with his peers or displayed eor behavior away from his desk.

We focused our analysis on T-pattern #1, the most complex pattern found to occur in half of the observation files, and T-pattern #129, which was the last of the 133 T-patterns containing eac and occurring in over half of the observation files (Figures 2 and 3). m1 appeared in all the branches of the tree diagrams featuring eac, showing the strong association between the two categories.

When we increased the statistical significance requirements (to a level of p < .005 and a minimum occurrence setting of 50% of the observational records), we found 15 T-patterns, but the connection between eac and m1 was retained (see T-pattern 1 with the strictest parameter requirements in Figure 4).

However, in order to explore synergies between the four methods, we were interested not only in determining the extent to which eac was statistically associated with certain categories but also in identifying the categories with which it was not associated. On observing the frequencies of the different event-types for the target student fdo1f (see Figure 5), we found that eac was only connected to ml. In other words, while performing the classroom task (eac), the student exhibited no connections with Initiation, Area (except for m1), or missing categories. Another interesting observation is the high distribution of eor (off-task behavior), as this reflects the motivational competition that exists between on-task and off-task behavior.

In this analysis the category eor was also closely associated with m1, indicating that the student did not move from his desk while alternating between on-task and off-task activities.

In the next section, we investigate events that triggered eor (e.g., eiep) using the other techniques with the ultimate aim of achieving valuable, complementary insights.

Lag sequential analysis

The first technique analyzed in this section is lag sequential analysis (Bakeman, 1978; Bakeman & Quera, 1996, 2011; Sackett, 1978, 1979; Bakeman & Gottman, 1986), which was performed using SDIS-GSEQ, GSEQ5, and HOISAN.

The SDIS-GSEQ algorithm compares conditional and unconditional probabilities related to behaviors that occur in the form of transition frequencies after the criterion behavior (or given behavior as it is known in SDIS-GSEQ). The given behavior is established according to the needs of each study. Considering its relevance in our study, the eac category was established as both the criterion and the conditional behavior. Adjusted residuals for the positive lags R1 to R5 were calculated using the binomial test and the correction proposed by Allison and Liker (1982).

Separate datasets were generated for each of the eight sessions. Data were entered using the timed-event data option (type IV data). The resulting SDS files were then converted to MDS files using the GSEQ compiler. Lag sequential analysis was performed using the GSQ file (Table 4).

Using the results of this analysis (OUT file), we identified the occurrence of both excitatory or activating behavioral patterns (Res_adjust > 1.96, for p < .05) and inhibitory behavioral patterns (Res_adjust < -1.96, forp < .05) (Table 1).

The high values obtained for eac as both the criterion and conditional behavior indicates that the task is self-sustained, which is further supported by the fact that it has a statistically significant inhibitory relationship with espr, espe, ejge, ejgr, eor, eirp, and eiep. As shown by Table 2 eiep, eirp, espr, espe, ejge, ejgr, and eor also had a significant inhibitory relationship with eac.

To explore the connection between eac and m1 using lag sequential analysis, we analyzed lag 0, which corresponds to concurrent events (Table 5).

As seen, the adjusted residual was 4.61, indicating a strongly significant association between the two categories. The negative connection between eor and m1 is also noteworthy, as it shows that off-task behavior also occurs away from the student's desk, complementing findings from the first analysis.

Trend analysis

Using the adjusted residuals obtained in the lag sequential analysis, we applied trend analysis to assess the extent to which the relationships connecting eac with the other categories varied across the different lags analyzed.

Using the adjusted residuals available for each of the positive lags, R1 to R5 (Table 3), linear regression equations were generated in STATGRAPHICS and the resulting trends were displayed as graphs. As mentioned above, the eac category was established as both the criterion (Table 6) and the conditional behavior (Table 7).

The regression equation linking eac (criterion behavior) to eor (conditional behavior) in Table 5 -eac / eor = -45.38 + 2.792*Lags-provides information on the association between the behaviors that complements the results of the Tpattern and lag sequential analyses. The existence of a negative value at lag 0 (concurrence) again shows that these behaviors do not occur simultaneously, but rather in alternation, as seen previously in the T-pattern analysis.

The regression equation linking eor (criterion behavior) and eac (conditional behavior) was eor / eac = -45.98 + 3.37*Lags. Again, the existence of a negative value at lag 0 shows that these two events cannot occur simultaneously, regardless of whether or not they occur in the same setting (m1).

The final analysis of the organization of class time, particularly with respect to the relationship between eac and the other categories, was performed using polar coordinate analysis.

Polar coordinate analysis

Polar coordinate analysis is a powerful data reduction technique that depicts complex relationships between a given category, known as a focal category, and other categories in the form of vectors. We applied the technique to show the network of relationships between eac (the focal behavior, which corresponds to the given behavior in lag sequential analysis) and the other categories in SOC-IS (conditional behaviors) to generate a map showing the various interactions between categories. The technique was proposed by Sackett (1980) based on previous work by Bakeman (1978) and was subsequently enriched by the incorporation of the concept of genuine retrospectivity (Anguera, 1997). Used in combination with SOC-IS, it has proven very useful in studies analyzing children's behavior in the classroom (Anguera, Espinosa, & Santoyo, 2002; Anguera, Santoyo, & Espinosa, 2003; Espinosa, Anguera, & Santoyo, 2004) and in other observational studies (Hernández-Mendo & Anguera, 1998; Anguera & Losada, 1999; Aragón, Lapresa, Arana, Anguera & Garzón, 2017; Castañer et al., 2016; Gorospe & Anguera, 2000; Herrero, 2000; López, Valero, Anguera & Díaz, 2016; Morillo, Reigal & Hernández-Mendo, 2015; Perea, Castellano, Alday & Hernández-Mendo, 2012). We believe that polar coordinate analysis would also be a valuable tool for analyzing the impact of self-regulation interventions in children with high rates of behavior transitions (Norris, 2016; Slattery, et al., 2016).

Following Cochran's proposal (1954), the prospective adjusted residuals (positive lags) and retrospective adjusted residuals (negative lags) generated in the lag sequential analysis were used, respectively, to compute prospective and retrospective Z_sum values, which in turn were used to calculate vector length and angle.

The resulting vectors are located in different quadrants, depending on their angle: 0-90^o corresponds to quadrant I, 91-180^o to quadrant II, 181-270^o to quadrant III, and 271-360^o to quadrant IV. Quadrants I and III are symmetric, while quadrants II and IV are asymmetric. Quadrants I and III correspond, respectively, to mutual activating and inhibitory interactions. Quadrant II corresponds to an inhibitory relationship between the criterion and the conditional behavior (or an activating relationship when the opposite direction is detected), while quadrant IV corresponds to an activating relationship between the criterion and the conditional behavior (or an inhibitory relationship for the opposite direction).

The use of polar coordinate analysis has been greatly facilitated by the inclusion of a dedicated feature in the software program, HOISAN. All the vector parameters were calculated and displayed as graphs using this program (version 1.6.3.2) (Hernández-Mendo, López-López, Castellano, Morales-Sánchez, & Pastrana, 2012).

The results of the polar coordinate analysis are shown in Table 8 and Figures 6 and 7.

In quadrant I, the only category that had a mutual activating relationship with eac was m1 (the corresponding vector had a length of 11.57 and an angle of 45.07^o). Given the asymmetric relationship that characterizes quadrant IV, in which the focal behavior has an activating effect on all other categories in the quadrant, the only categories it could affect were eal and ejp, for which the corresponding vectors had null values.

While the results of quadrants II (inhibitory focal behavior) and III (mutually inhibitory focal and conditional behavior) are not directly pertinent to our analysis, they are consistent with the results of the other techniques. The fact that eac had a significant inhibitory effect on the social categories in quadrants II (espr, eirp) and III (espe, esne, esnr, ejge, ejgr, eiep, eien, and eirn) shows an "incompatibility" that permits a greater understanding of how time is organized in the classroom.

 

Discussion

The aim of this study was to show how complementary observational methodology techniques can be used to analyze the distribution of classroom behavior from the perspective of a single child. Our results therefore should be contemplated within the context of a case study, i.e., they are not intended to be generalized to the broader population of students. The majority of studies in this area have based their findings on information collected through questionnaires and other sources of indirect data. Studies of the distribution of time in classroom settings can offer highly relevant insights into what actually occurs in the classroom and how this impacts student motivation, academic activities, and other relevant experiences. The main strength of the current study is that we analyzed behavioral patterns related to academic tasks performed by primary school children from an individual and group perspective.

Our specific aim was to use four complementary techniques to study relationships between on-task behavior and other interaction categories to analyze how students spend their time in the classroom. We focused on on-task behavior (the eac category) to address some of the gaps in knowledge on how classroom time is organized.

Considering, a priori, that it is difficult to interpret relationships between the eac category and other interaction categories using any of the four techniques-T-pattern detection, lag sequential analysis, trend analysis, and polar coordinate analysis-in isolation, we decided to use a combination of the techniques and assess the convergence between results.

Table 9 shows a side-by-side comparison of the results and helps to detect interpretative synergies.

The first finding of note is that according to the results of the T-pattern, lag sequential, and polar coordinate analyses, on-task behavior (eac) was not significantly associated with any other episodes and the only consistent connection observed was with m1.

Of particular note also is the statistically significant mutually inhibitory relationship between eac and both ejge (episode of group play in target child,) and ejgr (episode of group play in target peers). When the eac category was analyzed as the criterion behavior, both ejge and ejgr generated vectors located in quadrant III, indicating that play has a disruptive effect on on-task behavior. The adjusted residuals for lags R1 to R5 also exhibited an upward, albeit insignificant trend, indicating that the longer the play episode, the less likely the student is to start an academic task.

The espr (positive social interaction [prosocial behavior) by target child) and espe (positive social interaction in target peers) categories are also particularly relevant. Both exhibited a significant inhibitory relationship with eac, indicating that social interaction interferes with academic activity. In the case of espe, the relationship was also reciprocal, as the corresponding vector was located in quadrant III, unlike that of espr, which was located in quadrant II. The adjusted residuals for the positive lags R1 to R5 exhibited an upward trend, reflecting, in a statistically significant manner, the consolidation of this inhibitory relationship. The relationships between on-task behavior and social interaction (espr and espe) have important implications for education and the organization of classroom activities. The inclusion of categories that permit the identification of the person who initiates a social episode allowed us to confirm the motivational importance of actions initiated by the target child or his peers favoring off-task behavior. These aspects are evident in both the polar coordinate and T-pattern analyses.

Finally, the relationship between the eac category and both the eirp (positive social interaction initiated by peer) and the eiep category (positive social interaction initiated by target child) is also interesting. We observed a statistically significant inhibitory relationship between the two categories and eac as both the criterion and conditional behavior, highlighting the fact that initiation of social interaction distracts children from academic tasks. The eiep vector was located in quadrant III, while the eirp vector was located in quadrant II. The inhibitory nature of these categories was further supported in the trend analysis, as they exhibited an upward trend, regardless of whether eac was analyzed as the criterion or conditional behavior. Furthermore, in the case of eiep, the upward trend was statistically significant when eac was defined as the conditional behavior.

We have presented an extension of previous work by our group in the field of academic behavior (Santoyo & Anguera, 1992) that supports previous findings from observational methodology studies conducted in similar school settings (Santoyo et al., 2000, 2006). In particular, our study adds to the body of scientific literature on the distribution of classroom behavior that has focused on the time spent by teachers in getting their students started on a task. Time lost in achieving this logically has an effect on the time spent by students on academic tasks. One general limitation of studies in this area to date is that they have based their analyses on indirect data rather than on actual time spent on class-work or factors that influence on-task behavior. Future studies should address motivational factors in situ to shed light not only on the allocation of time by teachers in the classroom (Martinic, 2015) but also on the distribution of activities that form part of on- and off-task behavior during teacher-student and student-student interactions in the classroom (Santoyo, in press).

To conclude, the combined use of four complementary observational methodology techniques to explore in greater depth the distribution of classroom activities in a primary school setting shows that the methods converged in terms of essential information but also provided differential insights that, considered as a whole, add an interesting depth to the analysis. We believe that more studies should aim to enrich their analyses and interpretation of data by using complementary techniques.

 

References

1. Abramowitz, A.J., O'Leary, S.G., & Futtersak, M.W. (1988). The relative impact of longand short reprimands on children's off-task behavior in the classroom. Behavior Therapy, 19, 243-247.         [ Links ]

2. Allison, P.D. & Liker, J.K. (1982). Analyzing sequential categorical data on dyadic interaction: A comment on Gottman. Psychological Bulletin, 91(2), 393-403.         [ Links ]

3. Anguera, M.T. (1997). From prospective patterns in behavior to joint analysis with a retrospective perspective. En Colloque sur invitation «Méthodologie d'analyse des interactions sociales». París: Université de la Sorbonne.         [ Links ]

4. Anguera, M. T., Blanco, A. y Losada, J. L. (2001). Diseños observacionales, cuestión clave en el proceso de la metodología observacional (Observational designs, key issues in the observational methodology process). Metodología de las Ciencias del Comportamiento, 3(2), 135-160.         [ Links ]

5. Anguera, M.T., Espinosa, M.C. y Santoyo, C. (2002). Observación de la conducta interactiva en niños: Análisis de la intensidad interactiva diádica (Observation of interactive behavior in children). Metodología de las Ciencias del Comportamiento, vol. especial, 34-36.         [ Links ]

6. Anguera, M.T. y Losada, J.L. (1999) Reducción de datos en marcos de conducta mediante la técnica de coordenadas polares (Reduction of data in behavioral frameworks using the polar coordinate technique). En M.T. Anguera (Ed.), Observación de la conducta interactiva en situaciones naturales: Aplicaciones. Barcelona: E.U.B.         [ Links ]

7. Anguera, M.T., Santoyo, C. & Espinosa, M.C. (2003). Evaluating links intensity in social networks in a school context through observational designs. In R. García Mira, J.M. Sabucedo Cameselle & J. Romay Martínez (Eds.), Culture, Environmental Action and Sustainability (pp. 286-298). Göttingen: Hogrefe & Huber.         [ Links ]

8. Aragón, S., Lapresa, D., Arana, J., Anguera, M.T., & Garzón, B. (2017). An example of the informative potential of polar coordinate analysis: sprint tactics in elite 1500 m track events. Measurement in Physical Education and Exercise Science, 21(1), 26-33, DOI: 10.1080/1091367X.2016.1245192.         [ Links ]

9. Bakeman, R. (1978). Untangling streams of behavior: Sequential analysis of observation data. In G.P. Sackett (Ed.), ObservingBehavior, Vol. 2: Data collection and analysis methods. Baltimore, University of Park Press, pp. 63-78.         [ Links ]

10. Bakeman, R., & Gottman, J.M. (1986). Observing Interaction: An introduction to sequential analysis. New York: Cambridge University Press.         [ Links ]

11. Bakeman, R. y Quera, V. (1996). Análisis de la interacción. Análisis secuencialcon SDIS y GSEQ (Interaction analysis. Sequential analysis with SDIS and GSEQ). Madrid, Ra-Ma.         [ Links ]

12. Berk, L.E., & Landau, S. (1993). Private speech of learning disabled and normally achieving children in classroom academic and laboratory contexts. Child Development, 64, 556-571.         [ Links ]

13. Castañer, M., Barreira, D., Camerino, O., Anguera, M. T., Canton, A., and Hileno, R. (2016). Goal scoring in soccer: A polar coordinate analysis of motor skills used by Lionel Messi. Frontiersin Psychology, 7, 806. doi: 10.3389/fpsyg.2016.00806.         [ Links ]

14. Cochran, W.G. (1954). Some methods for strengthening the common χ² test. Biometrics, 10, 417-451.         [ Links ]

15. Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37-46.         [ Links ]

16. Cohen, J. (1968). Weighted kappa: Nominal scale agreement with provision for scaled disagreement of partial credit. Psychological Bulletin, 70, 213-220.         [ Links ]

17. Delgado, E.L.G. (2013). La ley de igualación, de la teoría al aula escolar: La relación entre la atención diferencial y la conducta dentro del aula (The matching law, from theory to the classroom: The relationship between differential attention and behavior in the classroom). Tesis de Licenciatura en Psicología. Universidad Nacional Autónoma de México.         [ Links ]

18. Elder, G.H., Pavalko, E.K., & Clipp, E.C. (1993). Working with archival data: Studying lives. Newbury Park: Sage Publications.         [ Links ]

19. Espinosa, M.C., Anguera, M.T. y Santoyo, C. (2004). Análisis jerárquico y secuencial de patrones sociales "rudimentarios" establecidos por niños pequeños (Hierarchical and sequential analysis of "rudimentary" social patterns established by young children). Metodología de las Ciencias del Comportamiento, Suplemento, 193-200.         [ Links ]

20. Espinosa, A.M.C., Blanco-Villaseñor, A., & Santoyo, V.C. (2006). El estudio del comportamiento en escenarios naturales: Observación de las interacciones sociales (Studying behavior in natural settings: Observation of social interactions). En C. Santoyo y M.C. Espinosa (Eds.). Desarrollo e interacción social: Teoríay métodos de investigación en contexto (pp. 45-78). México: UNAM/CONACYT.         [ Links ]

21. Gorospe, G. y Anguera, M.T. (2000). Modificación de la técnica clásica de coordenadas polares mediante un desarrollo distinto de la retrospectividad: Aplicación al tenis (Modification to the classic polar coordinate technique using a distinct concept of retrospectivity: An application to tennis). Psicothema, 12 (Supl. N^o 2), 279-282.         [ Links ]

22. Hernández Mendo, A. y Anguera, M.T. (1998). Análisis de coordenadas polares en el estudio de las diferencias individuales de la acción de juego (Polar coordinate analysis in the study of individual differences of the game action). En M.P. Sánchez y M.A. Quiroga (Coords.), Perspectivas actuales en la investigación psicológica de las diferencias individuales (pp. 84-88). Madrid: Centro de Estudios Ramón Areces.         [ Links ]

23. Hernández-Mendo, A., Castellano, J., Camerino, O., Jonsson, G., Blanco-Villaseñor, A., Lopes, A. y Anguera, M.T. (2014). Programas informáticos de registro, control de calidad del dato, y análisis de datos (Observational software, data quality control and data analysis). Revista de Psicología del Deporte, 23(1), 111-121.         [ Links ]

24. Hernández-Mendo, A., López-López, J. A., Castellano, J., Morales-Sánchez, V. y Pastrana, J. L. (2012). Hoisan 1.2: Programa informático para uso en metodología observacional (Hoisan 1.2: Software for observational methodology). Cuadernos de Psicología del Deporte, 12(1), 55-78.         [ Links ]

25. Herrero, M. L. (2000). Utilización de la técnica de coordenadas polares en el estudio de la interacción infantil en el marco escolar (Use of the technic of polar coordinates in the study of the children interaction in the school room). Psicothema, 12(2), 292-297.         [ Links ]

26. Herrero, M.L. y Pleguezuelos, C.S. (2008). Patrones de conducta interactiva en contexto escolar multicultural (Patterns of interactive behavior in a multicultural school setting). Psicothema, 20(4), 945-950.         [ Links ]

27. Heward, W.L. (1994). Three "low-tech" strategies for increasing the frequency of active student response during group instruction. En R. Gardner, D. Sainato, J. Cooper, T. Heron, W. Heward, W. Eshleman, & T. Grossi (Eds.), Behavior analysis in education: Focus on measurably superior instruction (pp. 161-172). Pacific Grove, Cal.: Brooks /Cole Publishing Co.         [ Links ]

28. Heller, M.S., & White, M.A. (1975). Rates of teacher verbal approval and disapproval to higher and lower ability classes. Journal of educational psychology, 67, 796-800.         [ Links ]

29. López, J., Valero, A., Anguera, M. T., & Díaz, A. (2016). Disruptive behavior among elementary students in physical education. Springer Plus, 5, 1154. doi: 10.1186/s40064-016-2764-6.         [ Links ]

30. Magnusson, M.S. (1996). Hidden real-time patterns in intra-and interindividual behavior: Description and detection. European Journal of Psychological Assessment, 12(2), 112-123.         [ Links ]

31. Magnusson, M.S. (2000). Discovering hidden time patterns in behavior: T-patterns and their detection. Behavior Research Methods, Instruments, & Computers, 32(1), 93-110.         [ Links ]

32. Magnusson, M.S. (2005). Understanding social interaction: Discovering hidden structure with model and algorithms. In L. Anolli, S. Duncan, M.S. Magnusson, & G. Riva (Eds.), The hidden structure of interactions: From neurons to culture patterns (pp. 4-20). Amsterdam: IOS Press.         [ Links ]

33. Martens, B. K. (1990). A context analysis of contingent teacher attention. Behavior Modification, 14, 138-156.         [ Links ]

34. Martens, B.K., Halperin, S., Rummel, J.R., & Kilpatrick, D. (1990). Matching theory applied to contingent teacher attention. Behavior Assessment, 12, 139-155.         [ Links ]

35. Martinic, S. (2015). El tiempo y el aprendizaje escolar: la experiencia de la extensión de la jornada escolar en Chile (Time and learning at school: the experience of lengthening the school day in Chile). Revista Brasileira de Educação, 20, 61, 479-499. http://dx.doi.org/10.1590/S1413-24782015206110.         [ Links ]

36. Morillo, J. P., Reigal, R. E., and Hernández-Mendo, A. (2015). Análisis del ataque posicional de balonmano playa masculino y femenino mediante coordenadas polares (Analysis of positional attack in beach handball male and female with polar coordinates). Revista Internacional de Ciencias del Deporte, 11, 226-244.         [ Links ]

37. Norris, L.A. (2016). Self-Regulation Strategies for Students with Disruptive Behavior Disorders. Culminating Projects in Special Education. Paper 7. http://repository.stcloudstate.edu/sped_etds.         [ Links ]

38. Organization for Economic Cooperation and Development (OECD) Teaching and Learning International Survey (TALIS) 2013 http://www.oecd.org/edu/school/Questionnaires%20TALIS%202013.pdf.         [ Links ]

39. Otero, T.L., and Haut, J.M. (2016). Differential effects of reinforcement on the selfmonitoring of on-task behavior. School Psychology Quarterly, 31(1), 91-103. doi: 10.1037/spq0000113.         [ Links ]

40. Perea, A., Castellano, J., Alday, S., and Hernández-Mendo, A. (2012). Analysis of behaviour in sports through Polar Coordinate Analysis with MATLAB. Quality & Quantity, 46, 1249-1260. doi: 10.1007/s11135-011-9435-z.         [ Links ]

41. Razo, P.A.E. (2015). Tiempo de aprender. El uso y organización del tiempo en las escuelas primarias en México (Time to learn. Use and organization of time in primary schools in Mexico). Memorias del Segundo Congreso latinoamericano de medición y evaluación educacional. Instituto nacional para la evaluación de la educación. 12-14 de marzo. Ciudad de México.         [ Links ]

42. Sackett, G.P. (Ed.) (1978). Observing Behavior, Vol. 2: Data collection and analysis methods, Baltimore: University of Park Press.         [ Links ]

43. Sackett, G.P. (1979). The lag sequential analysis of contingency and cyclicity on behavioral interaction research. In J.D. Osofsky (Ed.) Handbook of infant development, New York: Wiley, pp. 623-649.         [ Links ]

44. Sackett, G.P. (1980) Lag sequential analysis as a data reduction technique in social interaction research. In D.B. Sawin, R.C. Hawkins, L.O. Walker & J.H. Penticuff (Eds.). Exceptional infant. Psychosocial risks in infant-environment transactions (pp. 300-340). New York: Brunner/Mazel.         [ Links ]

45. Santoyo, C. (in press). Uso del tiempo en el aula: Una perspectiva observacional de la "organización" de patrones de comportamiento (Use of time in the classroom: An observational perspective of the "organization" of behavioral patterns). In C. Santoyo (Comp.) Investigación puente de Mecanismos Básicos de Toma de Decisiones Intrapersonales, Diádicas y Grupales. México: CONACYT/UNAM.         [ Links ]

46. Santoyo, C. (2010). Reflexiones conceptuales sobre la Persistencia académica: Aportaciones de un Enfoque de Síntesis (Conceptual reflections on academic persistence: Contributions from a synthesis-based approach). Revista Mexicana de Psicología Educativa, 1(1), 5-11.         [ Links ]

47. Santoyo, C. (2006). Persistencia y transiciones en escenarios naturales: Desarrollo, interferencia social e interrupción de patrones de comportamiento (Persistence and transitions in natural settings: Development, social interference and interruption of behavioral patterns). En C. Santoyo y C. Espinosa (Eds.), Desarrollo e Interacción Social: Teoría y métodos de investigación en contexto. (Pp. 181-212). México: UNAM/CONACYT.ISBN:970-32-2658-2.         [ Links ]

48. Santoyo, C. (2007). Estabilidad y cambio de patrones de comportamiento en escenarios naturales: Un estudio longitudinal en Coyoacán (Stability and change in behavioral patterns in natural settings: A longitudinal study in Coyoacán). México: UNAM/CONACYT.         [ Links ]

49. Santoyo, C., & Anguera, M.T. (1992). El hacinamiento como contexto: Estrategias metodológicas para su análisis (Crowding as context: methodological strategies for its analysis). Psicothema, 4, 551-569.         [ Links ]

50. Santoyo, C., & Colmenares, L. (2012). Investigación puente y de archivo: Implicaciones para el Estudio Longitudinal de Coyoacán (Bridge and archive studies: implications for the Longitudinal Coyoacán Study). En C. Santoyo (Coord.), Aristas y perspectivas múltiples de la investigación sobre desarrollo e interacción social. (Pp. 43-74). México: UNAM/CONACYT 57327.         [ Links ]

51. Santoyo, C., & Espinosa, M.C. (2006). Desarrollo e interacción social: Teoría y métodos de investigación en contexto (Development and social interaction: Theory and research methods in context). México: UNAM/CONACYT.         [ Links ]

52. Santoyo, C. Espinosa, M.C. & Bachá, M.G. (1994). Extensión del sistema de observación conductual de las interacciones sociales: calidad, dirección, contenido contexto y resolución (Extending the behavior observation system in social interaction contexts: quality, direction, content, context and resolution). Revista Mexicana de Psicología, 11, 55-68.         [ Links ]

53. Santoyo, C., Espinosa, M. C., & Bachá, G. (1996). Una estrategia para el análisis de la organización del comportamiento social (A strategy for analyzing the organization of social behavior). Revista Mexicana de Análisis de la Conducta, 22, 1, 79-93.         [ Links ]

54. Santoyo, C., Fabián, T.A., & Espinosa, M.C. (2000). Estabilidad y cambio de procesos de interferencia social: Un estudio longitudinal con niños de primaria (Stability and change in social interference processes: A longitudinal study with primary school children). Revista Mexicana de Análisis de la Conducta, 26, 299-317.         [ Links ]

55. Santoyo, V.C., Morales, Ch. S., Colmenares, V.L., & Figueroa, B.N. (2007). Organización del comportamiento en el aula: Transiciones, persistencia, interrupciones e interferencia social (The organization of classroom behavior: Transitions, persistence, interruptions, and social interference). En C. Santoyo (Editor). Estabilidad y cambio de patrones de comportamiento en escenarios naturales: Un estudio longitudinal en Coyoacán. (pp. 149-180). México: CONACYT 40242H/UNAM.         [ Links ]

56. Slattery, L., Crosland, K., & Iovannone, R. (2016). An Evaluation of a Self-Management: Intervention to Increase On-Task Behavior With Individuals Diagnosed With Attention-Deficit/Hyperactivity Disorder. Journal of Positive Behavior Interventions, 18, 3, 168-179. DOI: 10.1177/1098300715588282.         [ Links ]

57. Tarragó, R., Iglesias, X., Lapresa, D., & Anguera, M.T. (2016). Complementariedad entre las relaciones diacrónicas de los T-patterns y los patrones de conducta en acciones de esgrima de espada masculina de élite (Complementarity between diachronic T-pattern relationships and behavioral patterns in elite male fencing). Cuadernos de Psicología de Deporte, 16(1), 113-128.         [ Links ]

58. Tarragó, R., Iglesias, X., Lapresa, D., Anguera, M.T., Ruiz-Sanchis, L., & Arana, X. (in press). Analysis of diachronic relationships in successful and unsuccessful behaviors by world fencing champions using three complementary techniques. Anales de Psicología (in this monographic number).         [ Links ]

59. Torres, L.G.Y. (2012). Análisis de la interacción con el profesor y las actividades escolares en niños de primaria (Analysis of interactions with the teacher and school activities in primary school children). Tesis de Licenciatura en Psicología. Universidad Nacional Autónoma de México.         [ Links ]

 

 

Correspondence:
Carlos Santoyo V.,
Facultad de Psicología,
División de Investigación y Posgrado,
Edif. D, cub.120,
Av. Universidad 3304, Col.
Universidad Nacional Autónoma de México,
Deleg. Coyoacán, CP 04510,
Ciudad de México (México).
E-mail: carsan@unam.mx

Article received: 12-10-2016
revised: 29-11-2016
accepted: 28-02-2017