This project monitored first year student attendance in practical computing sessions across several courses at the School of Computing and Mathematical Sciences, University of South Wales (USW). Research has shown the correlation between attendance and success. We frequently observe a vicious cycle of poor attendance that leads to poor engagement and consequently poor results, the combination often self-perpetuating.
In order to break this cycle the project had three key objectives. The first was to establish an automated attendance monitoring system based on radio-frequency identification (RFID) and computer login data. The second was to identify how attendance data could be turned into information to improve the structure around year tutor support and course leadership. The final objective was to ensure that students could not only view their own individual data ‘picture’ but also interact and engage with the system.
Results are outlined for the first two objectives, showing that for the data to be of use, the teaching staff require accurate registers and a comparison for student attendance in other modules, to gain a full picture of the students’ attendance. Also year tutors / course leaders need to view the aggregated data for all students across all modules with drill-down functionality.
Research has shown a positive correlation between attendance and success as corroborated by Doyle et al. (2007), Duty (2012), Rhodes and Jinks (2005), and Teague and Corney (2011). Evidence in the literature also demonstrates that attendance and engagement can both be predictors of academic achievement (Bevitt et al., 2010; Blasco-Arcas et al., 2013). When examining attendance records Doyle et al. (2007, p. 131) found that ‘the greater the non-attendance the poorer the student performance’. Teague and Corney (2011, p. 1243) use attendance as one measure in ‘maintaining an engaged cohort’ claiming that attendance improved when students were engaged and ‘student failure rates’ plummeted. Kuh (2009, p. 683) defines engagement as:
‘The time and effort students devote to activities that are empirically linked to desired outcomes of college and what institutions do to induce students to participate in these activities’.
Therefore, if engagement is achieved by what we do to encourage students to participate in academic activities, (examples include, but are not limited to, student-staff interactions, induction and a supported learning environment), then engagement has some dependency on student attendance. Klem and Connell (2004, p. 267) state that ‘engagement is associated strongly with student attendance’ and therein lies the problem. Students do not attend. There are also issues more directly related to attendance that are caused, for example, by the difference between University and School: academic life is less prescriptive, less hand-holding and with a focus on proactive independent learning, more self-initiative is required and students need to manage their own time and responsibilities.
It is not to be implied that a student engages by merely attending but there seems to be ‘enough evidence across disciplines’ as written by Moore et al. (2008, p. 17) that ‘in order to optimise academic achievement’ students need to attend. This project proposes that attendance data can be used not only for the production of class registers but also in assisting with exchanges between staff and students. Staff support is ‘important to student engagement’ as cited by Klem and Connell (2004, p. 270). As a consequence of monitoring attendance the data can be manipulated to produce tabular presentations to assist when proactively approaching students who choose to miss classes. Self-registration and self-recording of non-attendance is an innovative form of interaction with students who can also provide their reasons for absence. Aptly used it could constitute a means of determining why students choose not to engage, giving possible insights for strategies to be put into place to improve attendance and success.
The computing subject area at USW has high first year student numbers, which causes issues when delivering applied knowledge to large groups with mixed abilities. This has in turn placed additional pressure on student support systems, including year tutors and course leaders. The Advice Centre for the School of Computing and Mathematical Sciences is a front-line desk for student support and queries and had previously monitored attendance, during lecture sessions only, using the electronic data recording system Uni-Nanny®. Since the demise of Uni-Nanny®, the Advice Centre gathers data from student interactions with Blackboard, Googlemail and GlamLife, the USW’s website / portal for students. This service cannot determine whether students are on-site or physically attending classes. Some, but not all lecturers monitor attendance with paper-based registers.
First year undergraduates at USW find programming difficult. Student expectations had not matched with experience and there was an unwillingness to engage in programming. Due to the lack of engagement, attendance for the programming units was poor and success rates disappointing. Programming consists of a large practical element and attendance is essential, as the student can demonstrate understanding of the concepts, be able to correct their misconceptions and also have better prospects of resolving problems. These practical sessions include summative assessments and the lecturers have the opportunity to assess the underpinning knowledge.
Attendance data is a vital early warning in identifying students potentially ‘at-risk’ as in a subject area that is highly cumulative, non-attendance can become increasingly dangerous. It is also an important indicator of engagement, (Krause et al. 2005), and we frequently observe at USW a vicious cycle of poor attendance that leads to poor engagement and consequently poor results. This combination is often self-perpetuating. Compounding the problem, the paucity of electronic attendance data in the School of Computing and Mathematical Sciences was hindering the School’s ability to ‘recognize and support students who are experiencing difficulties’, Bevitt et al. (2010, p. 1).
This project was undertaken to monitor student attendance for practical ‘high stakes’ classes to give a rich picture, not just for modules but also more importantly across courses. It had three key objectives:
- The first was to establish an automated attendance monitoring system based on radio-frequency identification (RFID) and computer login data.
- The second was to identify how attendance data could be turned into information to improve the structure around year tutor support and course leadership, providing a tool for academic staff to put interventions into place where points of weakness and challenges have been identified. The attendance data had other information mapped against it to determine whether other trends or patterns could be established with a view to engagement, retention and success.
- The final objective was to ensure that students could not only view their own individual data ‘picture’, that is, a graphical and / or tabular attendance record stratified by module or by week but also interact and engage with the system. The student may self-register and report non-attendance. They are also provided with a comments section should they wish to give a reason or explanation for their absence.
Attendance monitoring in ‘higher education is complex’, Bowen et al. (2006, p. 375) and for it to be meaningful the data has to be collected accurately and on a timely basis, without the onus being on the lecturer. This project proposes using RFID technology as an identification method to create an online student interactive system for recording and reporting student attendance whilst simultaneously supporting multiple users.
There are various technologies available today used for identification purposes; barcodes and Quick Response (QR) codes, RFID and its extension Near Field Communication (NFC). Barcode, QR code and NFC are short-distance technologies requiring a close touch to activate a reading. They were discounted for this project due to the high first year student numbers and possible congestion difficulties in registering for a class. RFID however can send information over fairly large distances, as shown in Table 1. It is a well-established technology, Juels (2006) that uses radio waves to read data stored on a tag that is attached to an object, an animal or a human being. Its function is for the automated identification and tracking of the object. The data is normally a unique serial number from which an identity can be established. Tags, readers and a computer system are the three main components of a RFID system.
Table 1 RFID frequency bands
|Frequency band||Description||Range (approx.)||Data speed||Use|
|<135 KHz||Low Frequency (LF)||10 cm||Low||Animal identification|
|13.56 MHz||High Frequency (HF)||1 m||Low to moderate||Smart cards|
|840-960 MHz||Ultra High Frequency (UHF)||1-12 m||Moderate to high||EAN, various standards. Car remote control keys / cards|
|2.45 GHz||Microwave Frequency||1-2 m||High||802.11 WLAN, Bluetooth standards|
Tag, smart label or transponder
Most RFID tags consist of a substrate material and two component parts. There is a tiny integrated circuit or chip that holds the information to be transmitted and it is this component that determines criteria such as the volume of data stored and the read / write speed. The tag also contains an antenna for receiving and transmitting a signal. The range of data transfer will depend on the shape, size and frequency of the antenna where the frequency governs not only the reading range but also speed and water penetration.
Passive, semi-passive tags and active tags
Passive tags are solely powered by the electromagnetic field of the reader. Semi-passive tags contain a battery to run the chip’s circuitry but use the reader for data transmission. Active tags contain their own battery power supply for the transfer of data and thus have the advantage of a greater range.
Dependent on usage, tags are produced in a variety of designs. In its simplest form, the tag consists of the chip and antenna simply being placed into a thin transparent film. Easily adapted to create adhesive labels and with additional durable coatings, the tags can then be made into cards onto which text and images can be printed. Tags may also have specialist casings applied for the insertion into living bodies.
To retrieve the data held by a tag, a reader or transceiver is required. The reader generates a weak radio signal, which is transmitted through the reader’s one or many antennae. The radio wave of the reader must be compatible with the tag. Whilst the reader is supplying the tag with energy the data stored in the memory of the tag can be read. The reader’s RFID middleware can interface with backend databases and information in a digital format via a Cat 6 Ethernet cable can be transferred to a computer system.
In 2013-2014 to encourage engagement and improve student success the USW introduced a new module in Procedural Programming for all first year HND and BSc (Hons) students. Although all 229 students participated in the same two-hour weekly lecture, 11 practical laboratory sessions were delivered to groups of between 9 and 33 students, see Table 2.
Wherever possible these smaller classes were taught by staff who would also teach them follow-on modules in the second half of the year in order to establish a better personal contact, almost on a personal tutor basis. There were five summative in-class test assessments and where the class groups were smaller, marking and individual feedback was performed immediately after the test and in class, on the premise that when students are provided with prompt feedback they learn more. In larger groups marking was performed separately and individual feedback provided in the following week. To further support students, Procedural Programming was chosen as the vehicle for the methodology and the foundation for the collection of attendance data and statistics along with another two core modules, see Table 3.
Table 2 Procedural Programming module groups
|Group||Staff||Number of students||Course|
|1||A||18||BSc Computer Forensics Group 1|
|2||B||22||BSc Computer Security|
|3||C||20||BSc Computer Science Group 1|
|4||D||25||BSc Computer Games Development Group 1|
|5||E||23||BSc Computer Games Development Group 2|
|6||F||2823||BSc Information Technology Management for BusinessBSc Software EngineeringBSc Computer Science Group 2|
|7||G||21||HND Computer Games Development|
|8||H||81||HND ComputingHND Information Communication Technology|
|9||J||99||BSc ComputingBSc Information Communication Technology|
|10||K||162||HND Computer Forensics and SecurityHND Software Engineering|
|11||M||17||BSc Computer Forensics Group 2|
Table 3 Modules identified for attendance monitoring using computer login data
|Module name||Number of students||Hours per week||Number of weeks||Number of groups||Paper registers|
|Procedural Programming||HND x 48BSc 176||2 hours||12 weeks||11||Yes|
|Information Engineering (Core Module)||BSc x 176||1 hour||24 weeks||8||First semester|
|Computer Systems and Network Technologies (Core Module)||BSc x 176||1 hour||24 weeks||8||No|
All students were made aware that the login data was being collected and their attendance monitored. This project is tracking movement, so to meet with ethical approval it is essential the students grant informed consent. This is explained to students during their first lecture when the tag cards are distributed. Information is provided on all aspects of participation, data use and they have the opportunity to ask questions and to receive answers. It is made clear that by signing for their tag card, students are giving their informed consent. However, it is also important that an opt-out is available. Should a student not wish to participate, their card is withheld and their student details removed from the system’s database as well as from the list containing student details that generates the login data. As a consequence they do not appear on any electronic registers or reports and will not be contacted or monitored as part of the project. A student may opt out at any point in time.
Login data collection
The login data, displaying date / time, student ID, room number and course was supplied by the IT User Services as a comma separated values file on a daily basis and provided a retrospective record from the previous day. It had been filtered for specific laboratories and only the computing first year students, see Figure 1. The login data in the first stage of the project was manipulated using spreadsheets to create typical registers, see Figure 2.
Figure 1 Filtered login data csv file
Figure 2 Paper-based register
These login data registers were made available for all the lecturers who taught on the Procedural Programming module, though they also maintained their own paper registers and comparisons were made to identify inconsistencies and errors. All login data registers for the Procedural Programming were monitored on a weekly basis and students who missed two or more consecutive sessions were contacted to ascertain the reason for their absence. Students were encouraged to notify staff of possible absences.
Whilst the same login data registers were maintained for the two core modules, they were not made available to the lecturers who taught Computer Systems and Network Technologies. These lecturers did not keep paper-based registers. The rationale for not offering the login data registers was simple; could the module be used as a control group to produce a comparison as to whether the lack of overt attendance monitoring made any significant difference to student attendance and engagement? The Advice Centre was kept informed when student attendance dropped or when supportive intervention was made.
RFID data collection
To gather attendance data electronically, an eight-port fixed Motorola FX9500 RFID reader (see Figure 5) and two UHF Advantenna-p12 antennae were installed into one laboratory. In an attempt to keep all the hardware unobtrusive it was placed into the space above the ceiling tiles approximately one metre inside the entrance door. The antennae were positioned at 90° to each other and inclined slightly to give the optimum field coverage.
In the first semester, three groups (amounting to 70 undergraduate students) attending the Information Engineering module were given a self-adhesive passive UHF Omni tag label and these were secured to their student ID cards, see Figure 3.
Figure 3 Student ID card with UHF tag label inserted below the student’s picture
The tag had a read range of up to four metres from the reader. Although each student was allocated a lanyard and holder, it was left to their discretion as to where they kept their ID card. The students participated on a voluntary basis only. Paper-based and login data registers were kept for these pilot groups and both compared against the tag label reads. Where tag labels had been attached to the student ID card, tag reads were disappointing. Unsurprisingly, none of the students chose to wear the lanyard and holder. Reads were made from those who placed the holder and ID card in the outside pocket of a rucksack. Most students however placed their ID cards inside a wallet, which was predominately kept for safe keeping about their person or in a handbag. The water content of the human body is enough to detune the antenna, resulting in reduced read range and therefore reduced tag reads. The antennae also captured too much data, with reads being made in the laboratory on the floor below.
Due to the poor results using tag labels, in the second semester the same students were given a credit card-sized UHF card with an embedded UHF chip and antenna, see Figure 4.
Figure 4 Students’ tag card
These cards had been printed with the student details and USW logo. The students were asked to have the card in their hand as they entered the laboratory. The tag card reads were analysed in the same manner as the tag labels and compared against the paper and login data registers.
The tag card reads were much more effective with records matching both paper and login data registers and gave consistent and accurate reads. The system was programmed for timetabled sessions expecting specific students for each session. Using an example for a session of one hour, the system would read and register tag cards ‘swiped’ at any time during the course of the full hour as well as 10 minutes before the hour. This 10 minute or buffer time was to accommodate for early students. The system was also programmed to use the login data as a backup should the student forget or lose their card and was set to the same parameters using a 10 minute buffer.
It was gratifying that all the students gave approval and agreed to pilot the system. They accepted the RFID attendance monitoring system and were not reticent in waving the tag cards upon entry to the room, showing interest in the interface to see whether they had been recorded as present.
Although the initial concept had been to keep the RFID antennae invisibly integrated by placing them above the ceiling tiles, it left the students slightly bemused as to whether they had brandished their tag card within the reading range. As the tag cards were more successful than the tag labels, the decision was taken to replace the two antennae in the ceiling with one UHF Advantenna-p11 antenna, placed on the wall immediately next to the entrance door at an accessible height for wheelchair users. The student then had a focal point for their tag reading.
Figure 5 (Motorola, 2014)
Figure 5 illustrates the system hardware installed into the laboratories. The first version of the system ‘R U Here?’, although in its infancy, was successfully operational. Objective 1, to establish an automated attendance monitoring system based on RFID and computer login data was achieved.
An accurate record was maintained of student attendance for all three modules throughout the academic year using login data and paper registers as controls. This was manipulated in a spreadsheet and used for comparative purposes when harvesting RFID reads. It was noticed that on some occasions the login data did not agree with paper based registers with students showing as absent. Further analysis of the login data highlighted the ambiguity. If students were timetabled in the same room but for different sessions they would only login to a computer once. The login data therefore would only reflect the student’s presence for the earlier session.
Objective 1 was to establish an automated attendance monitoring system. In its simplest format the attendance monitoring system just lists the tag reads. As the reader can read up to 200 tags per second and the range is fairly large, the list will contain many duplicate reads. The database will require a simple query for distinct tag reads and consequently student identity. Antennae are set with a suitable read range and face into the rooms. This is a measure to prevent spurious tag reads from students as they walk past in the corridor or who are present in the rooms above or below, rather than to avoid the system collecting data for which permission has not been granted. This issue has already been addressed by removing student details from the database, thereby not presenting on electronic registers irrespective of whether the student still carries a tag card.
There also conditions to be met. Students are expected to be in a particular laboratory on specific dates at certain times and therefore the system must anticipate these students. What the results showed in collecting the RFID reads and login data was that these conditions were not met if the student:
- Arrived earlier than 10 minutes to the hour.
- Was timetabled in the same room for different course sessions and did not swipe their card at the start of each distinct session.
- Session room and / or time changed.
- Migrated to another session for reasons such as friendship groups, preference of lecturer, absenteeism or simply to catch up.
- Did not swipe their card and did not login to a computer.
- Forgot or lost their card.
To maintain accuracy, several solutions were necessary to overcome these points:
- Use of the login data within the system to serve as a back up should a student forget to swipe or lose their card.
- The system must permit rooms and times to be altered if necessary.
- Specific software needed to be written to produce adjustable time buffers. If students arrived earlier than the default 10 minutes, their RFID reads and login data could be captured by setting the buffer to an earlier time.
- Student self-registration could further improve accuracy should a card malfunction or be forgotten, requiring the lecturer to verify to preserve integrity.
- Students attending non-timetabled sessions were recorded as additions to the register.
- Staff needed to be engaged in the project and remind students to swipe their cards. This again improved accuracy, especially when addressing the issue of students being timetabled in the same room but for different course sessions. Having logged in once it is unrealistic to expect them to repeat this process for each new session in the same room.
Objective 2 was to identify how attendance data could be turned into information to improve the structure around year tutor support and course leadership. This is a cyclical system of data gathering, analysis, identification of issues and intervention where needed. When monitoring the Procedural Programming module if attendance slipped, the data was acted upon immediately by email to the student, meeting them in person if the circumstances dictated and by liaising with the Advice Centre. The Advice Centre would not act on the information without referring to the student’s year tutor and this is where the first ‘gap’ emerged in the circle. The year tutors had no tools available to them to help put interventions into place to help students who were experiencing difficulties. The only electronic data, held by the Advice Centre was data from student interactions with Blackboard, Googlemail and GlamLife. At various stages of this project the author was contacted directly for data information to assist with student support enquiries.
Although all lecturers for the Procedural Programming module maintained paper-based registers, this is not a consistent practice across courses. These registers enable patterns of non-attendance to be established for students on a particular module, but give no information for the same students’ attendance in other modules.
Therefore if the system was to be of use there needed to be various ways of reporting the attendance data collected for improving the structure around year tutor support and course leadership.
The electronic or ‘Live View’ register (see Figure 6). This is a register for a session that is actually occurring or will happen. It is interactive and composed of three sections where students are present (student box has green handle), waiting (student box has grey handle) or absent (student box has red handle). As the reader scans the students’ tag cards they move from waiting to present. Should a student forget their card or choose to self-register, the lecturer can manually register or verify the student’s attendance by ticking the boxes. Students only show as absent when a session has finished. Those who have self-recorded a non-attendance also have a red handle.
The ‘Module Overview Report’ (see Figure 7). This gives the attendance for all students on a particular module week by week displaying a cumulative average attendance for that module. The module average is compared with the average attendance for other modules that the student is attending. It gives a much fuller picture of the students’ attendance patterns across their course.
The ‘Course Report’ (see Figure 8). This report gives an overview of each student’s attendance on a specific course.
The attendance data captured can provide not only individual student performance but also session and module performance across courses, offering a more coordinated approach. Modules are often shared across courses and due to student numbers modules may have many sessions per week, as illustrated in Table 2. Making this data available to year tutors will aid in deciding whether the students’ non-attendance requires support services to be put into place. It will effectively close the gap and complete the circle.
Figure 6 Live view register showing students that are present with green handles and ticked check boxes whereas students still to arrive are in the waiting section, grey coloured handles and unchecked boxes. Lecturers may manually register students by ticking check boxes.
Figure 7 Electronic graphical register showing each individual student attendance for a specific module compared to the student’s other modular attendance. Data is obscured to preserve anonymity
Figure 8 Course report that shows the attendance for each student on every module of their course
Objective 3 was to ensure that students could view their attendance data and be able to interact with the system. Version 1 of the ‘R U Here?’ system was successfully tested in one laboratory with 70 students during the second semester by simply collecting the raw tag reads and listing them. Version 2 was built during the summer when additional antennae were installed into another six laboratories in USW that use specialist software, all linking back via coaxial cables to the one reader. Bespoke software had been written to integrate RFID reads, login data and student self-registration. When the student initially opens the system the Home page gives a view of their timetable (see Figure 9). The student is able to interact with the system by:
- Uploading their image to personalise the register.
- Self-registering for a class should they forget or lose their tag card. They can record their presence manually into the system.
- Self-reporting their non-attendance.
- Using the comments tab for reason of absence.
- Viewing their attendance or individual data ‘picture’. Students can very often have a false impression about their attendance. Their view is a graphical and / or tabular attendance record stratified by module or by week.
Figure 9 Student view of the Home page
Click on a specific session and a window opens to permit students to self-register attendance and / or absence as well as leaving an explanation for their non-attendance. Comments are held in the Notes section on the menu bar. Selecting Attendance from the menu bar portrays a tabular view (see Figure 10).
Figure 10 Example of student attendance record for two modules
Findings of interest
During this project the students were fully aware that their attendance was being monitored and there were occasions in which they used absenteeism as a trigger for support, Bevitt et al. (2010). Students did not feel their absenteeism warranted talking to the Advice Centre but it was clear that they wanted their lecturer to be aware they were experiencing difficulties. Students were also often ‘unaware of who to contact’, Bowen et al. (2006, p.382) the author experiencing all of these issues on numerous occasions.
Verbal evaluation of the automated system has established that lecturers find it easy to use. They have further complemented the system with the incorporation of tablets. The tablet is quicker and easier than logging onto a computer in the laboratories and provides the opportunity for the ‘Live’ view of the electronic register (Figure 6) to be open at or before the start of the session. The tablet is almost a substitute for a paper based register, the lecturer can visually calculate from the electronic register’s live session how many students are present and a simple headcount proves / disproves its accuracy. Should a student not show as present it is merely a matter for the lecturer to manually register their attendance by touching the screen. The modular overview report (Figure 7) showed the attendance with a tabular view stratified by week and then compared by module. The lecturers found this a quick way of retrieving information to ascertain whether a student’s non-attendance was just a blip or indicated a trend towards possible disengagement. The system provides student email details for ease of contact as lecturer’s had requested that the system did not auto-generate a standard email, being of the opinion that students would simply discount the contact preferring a more personal approach. Soon after the first emails had been sent for non-attendance lecturers requested to have a means of recording student explanations of absence. This not only acted as a permanent record but also prevented a student being contacted repeatedly by many lecturers for the same absence.
Students indicated they would like to have a form of acknowledgement that the reader had registered their tag card, perhaps an audible beep or visually with a light flashing or changing colour.
Initial tentative findings suggest subtle differences in the attendance patterns between the different first year undergraduate students dependant on whether they are repeating, foundation or students in the first year of an undergraduate programme.
The system is more effective when staff are engaged with the system.
It is also vital that for an attendance monitoring system to be effective there must be accountability and responsibility from the users. Having the system is not enough without a structure to follow through having accessed the information that the data has supplied.
Anticipated causal relationship
The purpose of implementing the Procedural Programming module was to increase student attendance and improve success rates. It was expected to change the students’ attitude and behaviour in the subject area by building social capital with smaller classes and staff interactions, which are a strong form of engagement, (Bevitt et al. 2010). It is also putting into place a strategy to support students to be successful, (Thomas 2012b). This is made more effective if underpinned with the use of data to monitor student participation and attendance, identify students at risk and improve the structure around year tutor support and course leadership.
Whilst the attendance results for BSc Procedural Programming at 89% were higher when compared with the attendance results for the other two BSc core modules, Information Engineering 77% and Computer Systems & Network Technologies 71%, it is difficult to establish a clear cause-and-effect relationship between the work carried out and the results shown. This is the first year that the Procedural Programming module has been delivered and first year students are subjected to many more influencing factors than just those highlighted in the modules this project monitored. Next year the results will be stronger with not only the additional comparative data from monitoring all first year students’ attendance for all practical modules but also from the consequences of interventions that the academic staff have put into place where points of weakness and challenges have been identified.
This project intended to establish the feasibility of an automated attendance monitoring system and its worth in the correlation of improving student attendance and success. At the end of the first trial year we saw a significant improvement in both attendance and outcomes in terms of marks. It is, however, not possible at this moment in time to establish whether the two factors are linked by causation (higher attendance causing better outcomes), correlation (both improve as a result of the other supporting factors) or by a mix of the two. It could indeed be argued that the exact relationship between these factors can never be identified in a real-life setting in an ethically proper experimental setting.
The high level of attention paid to the Procedural Programming students during this project provided an insight into various issues such as the lack of electronic attendance data for the School of Computing and Mathematical Science students. It highlighted that the Advice Centre, a front-line desk for student support and queries, cannot proactively intervene when a student’s attendance falls as they refer back to year tutors for data. The year tutors do not have the tools to provide this information.
On the basis of this study it is argued that attendance monitoring can be the means of providing a tool for academic staff to proactively support students. Further rigorous analysis of the data after the future research phase has been implemented will determine whether attendance-monitoring data can be an influence on whether students persist or withdraw.
Blasco-Arcas, L., Buil, I., Hernández-Ortega, B., Sese, F. J. (2013) ‘Using clickers in class. The role of interactivity, active collaborative learning and engagement in learning performance’, Computers & Education, 62 pp. 102-110
Bevitt, D., Baldwin, C. and Calvert, J. (2010) ‘Intervening Early: Attendance and Performance Monitoring as a Trigger for First Year Support in the Biosciences’, Bioscience Education, 15. The Higher Education Academy Journals [Online]. Available at: http://journals.heacademy.ac.uk/doi/abs/10.3108/beej.15.4 (Accessed: 12 September 2014).
Bowen, E., Price, T., Lloyd, S. and Thomas, S. (2006) ‘Improving the quantity and quality of attendance to enhance student’, Journal of Further and Higher Education, 29(4) pp. 375-385
Doyle, L., O’Brien, F., Timmins, F., Tobin, G., O’Rourke, F. and Doherty, L. (2007) ‘An evaluation of an attendance system for undergraduate nursing students’, Nurse Education in Practice, 8(2) pp. 129-139
Duty, D. (2012). Seven Years at the Coal-Face: The Retention Phenomenon through the Lens of a Year Tutor, Published PhD thesis. University of Huddersfield
Juels, A. (2006) ‘RFID security and privacy: a research survey’, IEEE Journal on Selected Areas in Communications, 24(2) pp. 381-394
Keonn (2012) Available at: http://keonn.com/images/stories/pdf/Keonn-Advantenna-p12-Data_sheet.pdf (Accessed: 8 September 2014)
Klem, A. M. and Connell, J. P. (2004) ‘Relationships Matter: Linking Teacher Support to Student Engagement and Achievement’, Journal of School Health, 74(7) pp. 262-273
Krause, K. L., Hartley, R., James, R., and McInnis, C. (2005). The First Year Experience in Australian Universities: Findings from a Decade of National Studies: Department of Education Science and Training (DEST), Commonwealth of Australia.
Kuh, G. D. (2003). ‘What we’re learning about student engagement from NSSE’, Change, 35(2), pp. 24-32
Kuh, G. D. (2009). ‘What student affairs professionals need to know about student engagement’, Journal of College Student Development, 50(6), pp. 683-706
Mayer, R. E., Stull, A., DeLeeuw, K., Almeroth, K., Bimber, B., Chun, D., Bulger, M., Campbell, J., Knight, A., Zhang, H. (2009) ‘Clickers in college classrooms: Fostering learning with questioning methods in large lecture classes’, Contemporary Educational Psychology, 35(1) pp. 51-57
Moore, S., Armstrong, C., and Pearson, J. (2008). ‘Lecture absenteeism among students in higher education: a valuable route to understanding student motivation’, Journal of higher Education Policy and Management, 30(1), pp. 15-24
Motorola (2014) Available at: http://www.motorolasolutions.com/US-EN/Business+Product+and+Services/RFID/RFID+Readers/FX9500_US-EN (Accessed: 8 September 2014)
Rhodes, S., Jinks, A. (2005) ‘Personal tutors’ views of their role with pre-registration nursing students: An exploratory study’, Nurse Education Today, 25(5) pp. 390-397
Smith, E. M. and Beggs, B. J. (2003) ‘A new paradigm for maximizing student retention in higher education’, International Engineering Education Conference, Southampton.
Teague, D. & Corney, M. (2011). ‘Is anybody there? Bootstrapping attendance with engagement’, In G. Williams, P. Statham, N. Brown & B. Cleland (Eds.), Changing Demands, Changing Directions. Proceedings ascilite Hobart 2011, pp. 1239-1243
Thomas, L. (2012a) ‘What works? Facilitating an effective transition into higher education’, Widening Participation and Lifelong Learning, 14, pp. 4-24
Thomas, L. (2012b) ‘Building student engagement and belonging in Higher Education at a time of change: final report from the What Works? Student Retention & Success programme’ Available at https://www.heacademy.ac.uk/resources/detail/what-works-student-retention/What_Works_Summary_Report (Accessed: 8 September 2014)