an activity pattern for each student was plotted, using the SEm semester scores and last achieved SrL profiles. matriculation score (maT) is also pre- sented above the student’s plot. The information includes three intercon- nected main parameters of students’ learning: (1) student’s SrL profile and changes in this profile over time; (2) student’s synchronous activity mark (SEm) and changes in this mark over time; and (3) student’s achievements including their score in the final matriculation examination. The division formatted four groups (one for each SrL profile) presenting clear relation- ship and logical explanation for all variables.
5.3.1 General Interpretation of the Structure of Activity Patterns
Figure 5.3 demonstrates a typical activity pattern graph for each student in the course. It is important to emphasize that the a-SEm is not rep- resented in the activity patterns directly, but as a part of the student’s semester score.
G.a. is the name code of a student who joined the class in the second semester of the 11th grade. Components of G.a.’s activity pattern graphs
Figure 5.3 a typical activity pattern graph.
Downloaded from http://books.rsc.org/books/edited-volume/chapter-pdf/1746432/bk9781839165238-00049.pdf by RMIT University user on 06 February 2024
55 Personalized Support for Students Learning Chemistry Online
are: (1) the student’s name code (G.a.); (2) the student’s matriculation score (maT = 80); (3) year and semester (scale on x-axis: 10th [1–2], 11th [1–2], 12th [1–2]); (4) SEm score (scale on y-axis: 0–100%, represented by the small orange dots throughout the three-year course, during the summer breaks there are no dots since there was no SEm activity); and (5) semester scores, ranging from 30–100 on the y-axis. The color of the large dot represents the profile type of the student at the given point in time. In this example, the student was profiled as all-positive for the first semester of the 11th year, and as the fourth mixed type the year after.
Figures 5.4–5.10 show the activity patterns of various groups of students.
5.3.1.1 Very Active Students
high-achieving active students, who were also active during the synchronous lessons. Interestingly, very active students hold various SrL profiles, which for students G.a. and U.S. also changed over time. The matriculation scores for the students in this group did not fall below 80 (Figure 5.4).
Figure 5.4 activity patterns for very active students.
Figure 5.5 activity patterns for mildly active students—all-positive SrL profile.
Downloaded from http://books.rsc.org/books/edited-volume/chapter-pdf/1746432/bk9781839165238-00049.pdf by RMIT University user on 06 February 2024
Chapter 5 56
5.3.1.2 Mildly Active Students
The matriculation scores of students with this pattern were widely ranged.
we also found sub-trends within this group, according to the last exhibited SrL profile.
(a) Last SrL profile: all-positive SrL (Figure 5.5).
(b) Last SrL profile—all-negative SrL (Figure 5.6): students in this group were very low-functioning students who finally deteriorated to the all-negative SrL profile. all students in this group achieved low maT scores.
(c) Last SrL profile 3 (a mixed SrL profile, COn, TST > 0; STa, aTT < 0;
TmT, mOT ≈ 0) (Figure 5.7).
(d) Last SrL profile 4 (a mixed SrL profile, STa > 0; TmT, COn, TST < 0;
aTT, mOT ≈ 0) (Figure 5.8).
Figure 5.6 activity patterns for mildly active students—all-negative SrL profile.
Figure 5.7 activity patterns for mildly active students—last type 3 mixed SrL profile.
Downloaded from http://books.rsc.org/books/edited-volume/chapter-pdf/1746432/bk9781839165238-00049.pdf by RMIT University user on 06 February 2024
57 Personalized Support for Students Learning Chemistry Online
5.3.1.3 Non-active Students
In this group, we can see a very low SEm activity and medium to low maT scores. all students in this group had mixed SrL profiles with no changes along their three years of studies (Figure 5.9).
Figure 5.8 activity patterns for mildly active students—last type 4 mixed SrL profile.
Figure 5.9 activity patterns for non-active students.
Downloaded from http://books.rsc.org/books/edited-volume/chapter-pdf/1746432/bk9781839165238-00049.pdf by RMIT University user on 06 February 2024
Chapter 5 58
5.3.1.4 Diversely Active Students
In this group, we had students who became more active over time (positive trend) and students that decreased their activity over time (negative trend).
most students in this group became less active over time until they dropped out (Figure 5.10).
what have we learned from creating a personal activity pattern? The com- bination of students’ activity, SrL profile and semester achievements allows teachers to better interpret and understand student’s remote learning pro- cess, and address students’ personal needs for support.
however, can it be used to create a model that will allow prediction of a stu- dent’s success? Identifying such predictors and following them may increase success and lower the number of dropouts by allowing teachers to develop support strategies according to changes occurring in the student’s activity patterns, thus minimizing dropout.
we identified a clear relationship between the last-observed SrL profiles of the students and all other components: achievement group, maT score and activity pattern. all-positive last-observed SrL profile were high achiev- ers that were highly active during the lessons. Students with all-negative last- observed SrL profile, were low-achievers (except for one medium achiever) and low maT scoring students, all were mildly active students. Students with mixed last-observed SrL profiles (type 3 and 4) were mainly medium and high achievers with medium maT score students.
5.3.2 Success Predictors
Beyond the value of creating a methodology for tracking students’ activity profiles in a virtual environment, our main goal was to locate the most influ- ential factors that may serve as predictors, promoting student’s success and highlighting difficulties and needs, to support and minimize dropouts.
different predictors can be taken into consideration in order to locate the most influential ones that may promote student’s success.21 as discussed, there were quantitative measures that were related to success, although one Figure 5.10 activity patterns for diversely active students.
Downloaded from http://books.rsc.org/books/edited-volume/chapter-pdf/1746432/bk9781839165238-00049.pdf by RMIT University user on 06 February 2024
59 Personalized Support for Students Learning Chemistry Online
can argue that there are many types of success: for example, engagement, completion of the three-year course, leaving the course with a positive per- ception of chemistry, etc. In this study, students’ matriculation scores served as the outcome of the prediction. a success model was created, using mul- tiple linear regression to predict future maT scores, based on five indepen- dent variables: SrL profiles; achievement group; SEm; a-SEm; and activity patterns. Two promising success predictors were identified: involvement (expressed by SEm) and SrL profile type. The multiple linear regression results regarding the two influential predictors mean that a student pos- sessing an all positive SrL profile, can achieve a score of up to 38.53 points higher than a student possessing other SrL profiles. additionally, student’s maT scores may be increased by 24.25 points, for students who possessed a
‘mildly active’ activity pattern (over other activity patterns).