This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

The use of lecture capture in higher education is becoming increasingly widespread, with many instructors now providing digital videos of lecture content that can be used by students as learning resources in a variety of ways, including to catch up on material after a class absence. Despite accumulating research regarding the relationship between lecture capture and attendance, the nature of catch-up behavior following an absence has not been well characterized. This study measured attendance in relation to lecture video accesses to determine whether students catch up after missing a class, and if so, within what timeframe. Overall, it was found that 48% of absences were not associated with a corresponding lecture video access, and that when absences were caught up, the length of time taken to access the video was highly variable, with the time to the next exam being the likely determinant of when the video was viewed. Time taken to access a video was directly associated with deep learning approach score (as measured by the R-SPQ-2F). Males took significantly longer to view a corresponding lecture video after an absence than females, and missed significantly more classes than females. This study confirms that students use lecture capture variably, and that characteristics such as gender and learning approach influence lecture capture behavior including catch-up following an absence, a finding that is not unexpected given the diversity of students in higher education.

The term “lecture capture” describes the multimedia capturing of live lectures before an audience of students in higher education. While various elements of a lecture (such as instructor audio and video) can be captured, in this study the term ‘lecture capture’ is used to describe the use of digital technology to capture the instructor’s speech and their presentation slides, which were subsequently provided to students in digital video format. The use of lecture capture is becoming increasingly popular in undergraduate and professional education [

While there may be technological and pedagogical challenges associated with the use of lecture capture (for a review, see [

The relationship between lecture capture and student attendance has been the subject of considerable research attention over the last several years. Several studies suggest that there is only a minimal relationship between these variables, reporting that a majority of students still attend class despite lecture capture videos being available [

Research has shown that there are characteristics of students that may be important determinants of how lecture capture is used. Age, or academic level, is one such determinant. Chester

Regardless of whether the availability of lecture capture has a negative impact on attendance or not, it is well known by instructors in higher education that students regularly miss class. As characterized by Student Affairs at Penn State [

Subjects were enrolled in a second year undergraduate biochemistry course at the University of Guelph in Winter 2012 and Fall 2012. This undergraduate course provides students with the biochemical foundation for the study of human nutrition, exercise, and metabolism. The course covers aspects of biochemistry and metabolism that are critical to understanding human health and fitness. All students in Winter 2012 and Fall 2012 were invited to participate in the study. Student participation in Winter 2012 was 82.3% (28 out of the 34 total students, mean age 24.41 +/− 0.8 years, females = 11, males = 17 ) and in Fall 2012 participation was 76.5% (49 out of the 64 total students, mean age 19.75 +/− 0.14 years, females = 33, males = 16). Participating students received a 1% bonus mark on the final exam. In total, 77 students (78.6% of the total students between the two classes) participated in the study.

Lectures were captured using the iShowU software for Mac (Shiny White Box, New Zealand) and a Revolabs xTag USB wireless microphone (Revolabs, U.S). Videos were made of the live lecture, capturing both instructor audio and images on the instructor’s computer, which primarily showed

Video access was measured using trackable links in Courselink, the online learning management system. Absences were correlated with video accesses by measuring the time lapse between when the video was posted following an absence (described above as being from 0–2 days after lecture) and the date of the first time that a link was accessed. It should be noted that subsequent link accesses did sometimes occur; this study sought only to determine the time lapse to the first access.

The course was offered twice weekly on Tuesday and Thursday mornings, with 24 classes scheduled throughout the semester. The present study only considered attendance on days when lectures were given; after subtracting the dates of the two midterms, and five classes that were dedicated to group projects, a total of seventeen classes were included in the analysis. Attendance at each class for which lecture videos were available was monitored physically by collecting student signatures on a class list that was circulated mid-class. Signature lists were put into sealed envelopes after all attendees had the opportunity to sign, and were not viewed by the instructor until after submission of the final course grade. To determine whether absences were caught up by accessing a corresponding lecture capture video, each absence was treated as an independent data point. For the analyses of the length of time to view video, absences that were caught up by accessing the lecture video that corresponded to the missed lecture were similarly treated as independent data points and were grouped into the following categories: 1–7 days, 8–15 days, 16–25 days, 25 days or more. Numbering of days started from the day on which the lecture capture videos became available, which as previously mentioned, ranged from 0 (same day) to 2 days after the lecture. For the analyses of catch-up probability as related to student characteristics, probability was calculated using the following equation: number of classes caught up divided by total number of absences.

Student learning approach is commonly measured by the Revised Study Process Questionnaire 2-Factor (R-SPQ-2F) [

Final course marks were used as an indicator of academic performance.

Linear regression was conducted to determine the relationships between: (i) probability of catching up by accessing a lecture capture video and learning approach, age, and academic performance, (ii) the length of time to catch up (as divided into the previously described categories) and learning approach, age, and academic performance, (iv) learning approach and performance, and (v) learning approach and the number of classes missed. The Pearson correlation coefficient (r) was calculated for each association. The r value is considered to be the preferred index in a correlational design ([^{2} values and the unstandardized co-efficients for the y-intercept and learning approach score were also calculated. Chi-squared tests were conducted to investigate the relationships between gender and the length of time to catch up (as divided into the previously described categories). Effect size for the chi-squared analysis was determined by calculating the Phi coefficient. Unpaired, two-tailed t-tests were conducted to investigate the relationships between: (i) gender and the probability of whether absences were caught up (groups as yes or no), (ii) gender and the number of days (numerical, not categorical) taken to access a video after an absence, and (iii) gender and the number of total classes missed. Significance was indicated at a p-value of less than 0.05, and a trend was indicated by a p-value between 0.05 and 0.1. All analysis was done with SPSS version 20.

Catch up behavior was measured by tracking student attendance and subsequent accesses of links to video files in Courselink, the course management software. It was found that 48.4% of absences were not caught up by accessing the video that corresponded to the missed lecture. For absences that were caught up, these were divided into the categories previously described (

Catch-up behavior for each measured absence. This figure shows that 48.4% of absences were not caught up by accessing the video that corresponded to the missed lecture, while 11.5% of absences were caught up within 1–7 days of the video becoming available, 11.5% were caught up within 8–15 days, 10.2% were caught up within 16–25 days, and 18.2% were caught up after more than 25 days.

When measuring the relationships between the length of time to catch up and student characteristics, it was found that there was a significant relationship between deep approach score and the categories of number of days to catch up as measured by linear regression. A higher deep approach score was directly related to the category number of days to catch up by watching a video after missing a class (^{2} = 0.057) as represented by the equation [y (category of number of days to catch up) = 2.439 + 0.42 (deep learning approach score),

Males missed significantly (

Box and whisker plot showing the relationships between deep learning approach score (as measured by the R-SPQ-2F) and the length of time taken to access videos after an absence. There was a significant direct relationship between the variables (^{th} percentile and 25^{th} percentile respectively. The ends of the whiskers represent the minimum and maximum of the data.

Bar graph showing the length of time taken to access videos after an absence as a function of gender. There was a non-significant (

The probability of catching up was determined for each student by taking the total number of absences that were caught up, and dividing this by the total number of absences plus the total number of absences that were caught up. There were no significant differences (

Gender and probability of catching up from a missed class by watching a video. Error bars represent +/− SEM.

The pattern of video accesses across time for the Fall 2012 class, as well as the dates of the midterms and final exam, illustrate that the peaks of accesses parallel the dates of the examinations (

This figure shows the total number of videos viewed across the Fall 2012 semester. Examinations were held on 2 October 2012, 1 November 2012, and 8 December 2012.

Academic performance was significantly associated with deep learning approach score (^{2} = 0.101). There were no significant relationships between academic performance and surface approach score, gender, or age.

Although considerable research has investigated the relationship between lecture capture and attendance, whether students that miss a class actually catch-up by accessing the lecture video for the missed class, and the timeframe within which catch-up occurs, has not been well characterized. The objective of this study was to explore the catch-up behavior of undergraduate science students, and to determine whether behavior was influenced by student characteristics such as learning approach, age, and gender. To our knowledge, this is the first study to explore physical attendance in relation to the length of time taken to view a lecture video after an absence, and to evaluate the relationships between length of time to catch up and individual student characteristics that have been previously associated with lecture capture behavior.

The main finding of this investigation was that nearly half of student absences were not caught up by accessing the corresponding lecture video. Of course, this finding does not suggest that students do not catch up by some other means, such as borrowing another student’s notes or reviewing the slides from the missed lecture. Importantly, we did not find that academic performance was associated with the probability of catching up from an absence by accessing a corresponding lecture video, suggesting that the missed content was, in fact, addressed somehow by absent students. And, it should also be noted that accessing the video—which in this study, meant accessing a link that brought the students to the streaming video server—does not necessarily mean that students actually watched the video, or watched the video in full. However, these results are a strong indicator that students frequently do not access the only resource that will ensure full exposure to the material covered in lecture. Our findings are comparable to those of von Konski

The other main finding of this investigation was that for the 52% of absences that were caught up by viewing a corresponding lecture video, the time taken to access the video was highly variable, dispersing between the categories of catch up between 1–7 days, 8–15 days, 16–25 days, and greater than 25 days. When looking at the pattern of video accesses in this study, we observed that it peaks in parallel with the dates of each examination. This is consistent with numerous studies that report increased video accesses near the time of exams, and which indicate revision for exams as the primary motivator for using lecture capture videos [

Although not associated with academic performance, the length of time taken to catch up was significantly influenced by student approach to learning. Deep approach score, as measured by the R-SPQ-2F, was directly associated with the time to catch-up; that is, an increase in deep learning approach score was associated with an increase in the length of time taken to catch up by accessing a lecture video. This finding is somewhat surprising, since a deep learning approach is characterized by understanding, elaboration, analysis, and finding meaning from material [

The length of time taken to catch up was also significantly influenced by gender. Males took an average of 25 days to catch up from a missed class by watching a video, whereas females took an average of 16 days to catch up. The probability of catching up by accessing a lecture video after an absence was also associated with gender, with males being more likely to catch up than females. However, it should be noted that probability was calculated based on absences that were caught up as a function of total absences, and males missed significantly more classes than females. Chester

This study is limited by its small sample size, although the size is what allowed for effective measurement of physical attendance. The combined number of subjects across both classes was 77, which confers reasonable validity to the findings. It is very difficult to measure individual attendance in undergraduate classes, which is why most studies rely on self-reported, rather than actual, attendance. Additionally, there are characteristics of the program of study of the student participants in this study that may have influenced their attendance patterns. The program has several classes with mandatory attendance, which may encourage attendance across other classes as well. Also, the program has a heavy course load, which could constrain time and reduce the likelihood of students catching up from a missed class by accessing a corresponding lecture video. Although students were told that attendance sheets would not be viewed by the instructor until the end of the semester, it is known that attendance is higher in classes where attendance is recorded [^{nd} year course, the age dispersion may not have been sufficiently broad to investigate this relationship. Age was included as a variable in the analysis since the Winter 2012 class included students in a college transfer bridging program, whose mean age was almost 5 years higher than the Fall 2012 class, but these subjects comprised only 36% of total subjects. Future research should investigate the relationship between age and catch-up behavior using a bigger sample size across different academic levels.

Lecture capture is becoming increasingly popular in higher education, and the interest in this technology is projected to intensify in the near future. For this reason, it is important that educators characterize how lecture capture is used by students. This study found that a significant proportion of students do not catch up by accessing a lecture video after an absence, and that when they do, the length of time taken to access the video is likely determined by the time to the next exam. Whether the findings of the present study are interpreted negatively is intricately linked to the question of whether providing lecture capture to students results in a decrease in attendance, and to the nature of the relationships between lecture capture, attendance, and performance. If lecture capture does discourage attendance, then the observation that absences are often not caught up by accessing a corresponding lecture video, or are not caught up within a short timeframe, could be construed negatively; a similar impression would be valid if it were found that the probability of catching up, or the length of time taken to catch up, was associated with student performance. However, this interpretation is not necessarily supported by the literature, which mostly shows that attendance only changes minimally in response to provision of lecture capture [

The authors would like to thank Michelle Edwards at the University of Guelph for her valuable assistance with statistical analysis.

The authors declare no conflict of interest.