1.  For the Product Review

The numbering from the original chapter was:

Section 2.1 - the taxonomy of handsets

Section 2.2 - the taxonomy of software

Section 2.3 - actual systems

Needs editing

Everything got messed up in translation from pdf.

Might try it again from the original tex. Should be much easier.

2.  Chapter 2 - Hardware and Software

This chapter will briefly introduce the characteristics of currently available hardware and software for use as voting technology. The general idea of using handsets to gather data from an audience is found in all of these systems but each has some amount of variation from the rest. Before examining the specifics of some of the current systems in Section 2.3, the general variation that can be found across the different systems is discussed. This variation can be divided into two areas. Firstly, there is variation in the hardware and this is explored in Section 2.1 and tabulated against the specifics of current systems in Appendix A. Secondly, there is variation in the software and its features. This variation is investigated in Section 2.2 and, again, a comparison with current systems can be found in Appendix B.

2.1  Hardware Characteristics

There are many types of handset/receiver systems available. Variation in the hardware tends to be centred on a balance between cost and features. The important variations are detailed below, including the transmission method, the buttons on a handset and the provision of a display for feedback.

Transmission

The method by which data is sent from transmitter hardware to receiver hardware affects the possible function and the speed at which data can be gathered. Two different variables of the hardware design are worth considering: Radio Frequency vs. Infra-red and One vs. Two way communication.

Radio Frequency vs. Infra-red

The use of Infra-Red (IR) is thought of as less reliable than Radio Frequency (RF) as IR is considered ‘line-of-sight’ i.e. the handset must be in direct sight of the receiver. In an audience situation, this can be important if there are many heads and bodies between the transmitter and the receiver, or where a room has an unusual configuration. The handset systems however use high-powered IR LEDs and this means line-of-sight is not really an appropriate basis for analysis since signals will be reflected by most surfaces, but some consideration of physical objects interfering with the signal must be allowed. RF can be considered free from this problem.

The signal distance for IR is effectively less than for RF: most IR systems will work reasonably up to 20 metres whereas RF would certainly be over 50 metres. For most environments in which the handsets would be useful (certainly in HE), the 20m limit is more than sufficient if line-of-sight is available. On balance, it would seem that RF-based systems have the upper hand but the use of RF entails a much more complicated system: RF transmissions are limited to specific frequencies and tend not to be isolated to one room. This means the hardware must cope with interference from other sources operating on the same band, and this includes nearby class rooms using the same technology. These extra complications tend to increase the price, so RF devices are more expensive than those based on IR.

One or Two Way Communication

The transmission method can also be classified by whether transmission is either one-way or two-way. In one-way systems, the transmitter has no certainty that a signal has been received, whereas, with two-way communication, the receiver can send an acknowledgement signal and the transmitter can continue to resend until an acknowledgement is received.

Connected or Disconnected

Some newer, RF-based handsets can be used in what the manufacturer terms homework mode. The student is given a set of questions to answer and can enter these into the memory of the handset. The lecturer can then ask the students to submit their answers to all the questions in one go. This would allow students to take away the questions and handset and work on their answers in their own time. Then, at some appropriate time during a subsequent lecture, submit their answers. The answers themselves can often be numeric (either for MCQs or any other number-based answer) or sometimes, short text answers.

This operation is similar to the ability of other handsets to provide multi-digit responses as one transmission: they allow some operations to be carried out in a disconnected fashion. Most of the one-way devices require every button press to be transmitted and then decoded by the receiving computer: therefore, no such disconnected operation is possible.

Buttons

Handsets systems also vary in the number and function of their buttons. All systems have a set of buttons that represent the different answers a user would wish to submit, but some systems also provide special function buttons. The buttons that represent the submission of an answer vary across the systems in two respects. Firstly, the buttons either represent numeric or alphabetic digits and, secondly, the number of buttons varies. The use of alphabetic digits is often supplemental (i.e. a separate set of buttons) to numeric digits or as an overlay to numeric buttons (i.e. where button ‘1’ = ‘a’, ‘2’ = ‘b’, etc). Full alphabetic keyboards are rare and are currently limited to general purpose devices (i.e. systems based on PDAs or laptops). The provision of a full set of the ten digits, ‘0’ to ‘9’, is fairly standard, although some older systems provided less than ten buttons. These older systems may have followed understood guidelines for limiting the number of answer options in MCQs. The provision of the full ten digits is important in allowing a flexible use of the EVS as the handset can be used for more general data entry. For instance, the submission of a numeric value is possible, either as an integer or as a decimal where a decimal point can also be entered. A decimal point button, however, falls into the category of special function buttons, which can be identified simply by the fact that they do not represent a single digit answer. This can include, as well as a decimal point button, a negative sign (allowing the integers and decimals to be negatives) and buttons for navigation and mode change or modifier buttons.

Navigation buttons tend to be of the form ‘Next’, ‘Previous’ and ‘Enter’ though the naming varies. Mode change usually allows access to internal set up menus, which are examined (and altered) with the navigation and answer buttons.

Modifier buttons allow regular, single digit answers to be submitted with an extra data value attached representing, for instance, the voter’s confidence in his or her submitted answer. This is usually achieved by first pressing the modifier button and then the answer button. Modifier buttons are distinct from either answer buttons or the other special function buttons as they are a simple method of expanding the number of buttons available to a user whilst only adding a single extra button. For instance, a twenty digit handset could be considered identical to a handset with ten digit buttons and one modifier button: the first set of ten buttons is accessed by pressing an answer option button without pressing the modifier button, whilst the second set is accessed by pressing the modifier button before the answer button.

Display

An important aspect of the user experience of EVS is the feedback provided by the handset about the submission of an answer and its reception. Few systems provide no (visual) feedback and a range of methods can be found in current technology, from a single LED, an LED capable of multiple colours, multiple LEDs, a single line digit display, usually an LCD, or a display allowing multiple lines of alphanumeric characters, again usually an LCD. The physical hardware and the functionality of each system tend to dictate the complexity of the feedback mechanism: the more complicated the feedback needs to be, the more complicated the display mechanism also needs to be.

The simplest feedback is the use of LEDs to indicate that the handset is powered on and working and usually also to communicate that an answer is being sent (either by lighting or by flashing). Further, with two-way communication, the LEDs can be used to indicate that a vote has been properly received and acknowledged.

Where the handset can be used in a disconnected fashion as described in section 2.1.1, the use of at least a single line of digits is essential to allow users to clearly see the current state of their answers or operations before submitting them. The range of possible uses of particular handsets is vastly increased by the ability to display multiple lines of digits or alphabetic characters.

The use of RF technology usually dictates the need for a line of alphanumeric characters since the handset needs to be far more complex to cope with various problems of using radio transmission, not least of which is its ability to pass through solid objects. Unlike IR-based systems, the possibility exists for signals to be received by different systems in adjacent rooms. One solution to this particular problem, that demonstrates the need for a complex display, is for each handset to associate itself with a particular receiving system. This would require the user to navigate through a list of available receiving systems and select a particular one with which to communicate. The selection of a particular receiver is made easier where the list consists of easily understandable labels for each system (i.e. an alphanumeric string) and this would require a display capable of, at least, one line of alphanumeric characters.

Number of Handsets

Different systems seem to have different intended audiences and this is often reflected in a small upper limit being placed on the number of handsets that can be used in one class. However, this is rarer than a limit being placed on the number of handsets that can be used with a single receiver, and for larger audiences the addition of extra receivers is easily achieved.

PC connection

The connection to a standard PC or Mac laptop is required. There are essentially two methods by which this is currently achieved.

Firstly (both in simplicity and historically), some systems use a standard PC style nine pin serial port. The second method, the use of a USB port, is becoming more common.

The serial port is falling out of favour with computer manufacturers, as more and more devices need functionality beyond those provided by a standard serial port. It is a rare thing to find a desktop machine without a serial port but manufacturers have been producing laptops for some time that only have USB ports.

This is not a complete barrier to using serial-port-based devices as serial-to-USB adapters are readily available and relatively cheap. As Mac laptops never offered standard PC-style ports, this was always the method for these users (though older Mac serial ports can be adapted to simulate PC-style serial ports). Using these adapters has generally been successful, but it does add an extra level of complexity and potential point of failure.

The USB port is essentially faster and USB devices (i.e. not ones using a USB adapter) can draw power from the USB port itself. This reduces the need for extra power packs and for an available connection to the mains electricity, making the whole system much more portable.

Other aspects of interest

The batteries used by different handsets vary from a single AAA battery to a 9v battery. This is not a serious concern but does affect both the weight and the cost (in replacement batteries).

The battery life of each handset system is slightly different but a major division between the RF and IR handsets seems to exist: it is not unusual for IR handsets to have a battery life of a couple of years whilst the batteries for RF handsets are advertised as only lasting one academic year.

2.2  Software Characteristics

There is no standardisation of either hardware or software for voting systems (yet). This means that, generally, each hardware manufacturer supplies their own software, although third-party systems are also available which support more than one type of hardware.

The software systems differ as much as the hardware, but their main role is as a collection application: they collect and decode data representing button presses on handsets. Some features, such as graphing, are provided by all systems, although sometimes only as a secondary feature. Other features are peculiar to particular systems.

Whilst the specifics of current systems are discussed in subsequent sections of this chapter, the general variation in the software is discussed below. The major themes of variation are presented, including the means to prepare and present use of the system, possible question types and the display and management of data. A table is presented in Appendix B that shows a comparison of current software systems within this taxonomy.

Preparation and Presentation

The natural environment for the use of handsets would seem to be tied to the presentation of some other information. At the simplest level, the question and possible answers must be communicated to students, but this may not necessarily be in text: Answers can often take the form of pictures or diagrams. In most settings however, the handset use occurs in conjunction with information not directly related to the form of the question and answers and some allowance for this is generally seen. In this setting the natural companion to the collection application would be a standard presentation application, such as Microsoft PowerPoint or Apple KeyNote, though flexibility is required as other presentation systems may be used or the presentation of information may occur outside the computer realm (i.e. verbally, on OHP or blackboard etc). The range of ways that collection can be supplemented with other information includes:

• an expectation that the user should manually switch from a presentation (or other) application to the collection application;
• an automation of the switch at the appropriate moments;
• the provision of a free-text area of the screen;
• an internal presentation function of the collection application;
• integration with PowerPoint.

Ad Hoc use

The standard method of use of handsets (seen in most software systems) could be titled ‘ad hoc’ use, but it is really typified by needing no or little preparation of the handset systems software.

Obviously, the question and possible answers need to be communicated to voters but at its simplest, this is not done within the software component of the handset system. It can be either verbally (i.e. “Vote 1 for ... and 2 for ...”), written on a visible surface (i.e. blackboard, OHP, etc) or using a secondary display system on the computer (PowerPoint, Word, etc). From the software point of view, there is no preset ordering of questions and no set up of properties of collecting data for specific questions.

Internal Presentation Facility

A step beyond ad hoc use is achieved by the preparation of the ordering of questions (and their associated text) alongside any other content that is to be used or presented during the EVS use. The first way in which this can be done is where the collection application provides some internal presentation facilities. At its simplest, this does what was stated above: the ordering and/or the parameters of questions (i.e. the text, number of answer options) can be prepared in advance. This allows the presenter (i.e. lecturer) to be focused, during an EVS use, on the presented content rather than on the technology use: there is no need to remember the ordering of questions and the question text can appear on the same screen that is used to display the summarised results.

Internal presentation facilities are usually not as rich in features as standard presentation systems and require the user to learn how to use them. There would seem to be a balance to be struck between the complexity and, therefore, the user learning required and the features provided. An example of a fairly broad but simple method of providing a form of presentation is to allow the use of a series of picture files as a slideshow. This allows the user to build the ordering and content of their questions in anyway they like, as long as it can be reduced to a series of pictures.

PowerPoint Integration

The software developed as part of this thesis provides integration with PowerPoint (as discussed in Section 4.1.1) and therefore some fuller explanation is needed of how this works in other current software systems.

All systems that integrate with PowerPoint allow the user to mark particular slides in a presentation with question use. The difference among them is seen in what happens, during a slide show, when a marked slide is reached.

The first method to consider is where the collection application, which has been previously running in the background, becomes the front most application (i.e. sits on top of the PowerPoint slide show) at the point where a marked slide is reached in the slide show. The user (lecturer) has to then interact with the collection application until all the question data has been gathered. The lecturer must remain within the collection application while the graph is displayed before causing a switch back to the PowerPoint slide show.

The second method directly manipulates the content of the marked Power- Point slide. The collection application runs in the background and when the slide is reached, the data collection is started (either automatically or by the user clicking a button within the slide show). If a public display of each button press is needed (i.e. the handset number or the total number of respondents) this is drawn directly onto the PowerPoint slide. When the collection is finished, the graph is also drawn directly onto the slide. All of this occurs without interrupting or obscuring the PowerPoint slideshow.

The first method allows much greater tailoring of the display to the specifics of the data collection which often cannot be easily achieved, and is ultimately slower in the second method. However, the second method usually allows the user greater flexibility in what appears on screen at the time of collection (for instance the display of other information: text, pictures, movies etc). It also provides a more seamless visual environment: only one application is ever used.

Styles of Question

Multiple Choice Questions The simplest EVS use is based on Multiple Choice Questions (MCQ): each audience member has to pick exactly one option from a (short) list of possible answers. There are difficulties with the use of MCQs, however, and their use is not always desirable or appropriate. Recommendations for the form of MCQs suggest a small number of answer options, which limits the questions that can be used. Students tend to develop strategies for quickly eliminating answers that are obviously wrong: the testing may be of students’ ability to answer MCQs rather than their understanding of the content of the question. Different ways of presenting the question and answers, whilst still fundamentally working with multiple choice, have been found and applied successfully, for example, extended matching questions or pathway MCQ (Dugdale, 1998). These different presentation techniques overcome some of the inherent problems of MCQs.

However, when handset systems are seen as a data entry device in the hand of each student, a much greater flexibility presents itself, and current software systems facilitate this flexibility in a range of ways. In some systems, it is up to the user (lecturer) to reinterpret the raw button presses to have some new meaning. As discussed in section 2.2.6, this is not always possible in a live situation and must be performed afterwards. This is perhaps not as big a problem as the fact that, with MCQs, usually only the last button press from each student is retained. Some uses of handsets require more than one button press per question and without the collection software explicitly allowing these types of uses, there is little or no chance of achieving them. Some of these possibilities are discussed later, in Chapter 4. Discussed below are some of the features seen in current software, which behave differently from standard multiple choice use.

Multiple Digit

Questions that are more complex can be asked by allowing the use of multiple digit questions. This is still limited to numeric questions but not just to a number representing a choice but also to the input of a numerical value as the answer to questions such as “What year?”; “How old?”; “How many?”. This idea of a sequence of digits is essential to all of the following styles of use. Therefore, the provision of this in the software is a step toward providing more flexible use of handset technology.

Newer systems do allow the use of short text answers but it is unclear how in real time this could be analysed or summarised.

Multiple Answer

It is possible, given the idea of the multi-digit question above to extend this to allow combinations of answers to be entered. A question like “place these objects in order:” with a number of answers is achievable. This allows more discriminating questions and allows opinion questions of the form “from this list, pick your top 3”. The summation of this type of question is generally based around the ordering question and presents the number of people submitting a particular ordering. This is perhaps not appropriate to the pick-your-top-3 style question and little facility for displaying this sort of information is seen in the available software systems.

Multiple Questions

In most uses of these systems, button presses are answers to a single question, but some systems allow students to answer more than one question during a single data collection. The important consideration for the usability of this feature is how navigation between questions is achieved. That is, how does one select which question to answer and how is this distinguished from actually answering the question?

If the receiving computer is to decode each button press as an answer or as a navigation function, it needs to store the current question being answered and any previous answers given, for each student. Some display of the current state for each student must also be displayed on the laptop. In a two-way system, this is easier as there is a guarantee that each button press will have an effect. With one-way transmitters the task is not impossible but can be confusing and even frustrating to the student: each button press has to be sent and then a wait ensues to see if it causes the appropriate change at the receiving computer. This is further complicated in hardware systems that are not designed for navigation (i.e. that don’t have a specific set of navigation buttons). Further discussion of how this can be achieved are discussed in Section 4.4.2 and details of its practical application are given in 5.1.8

In some two-way systems with LCD screens, answering multiple questions, particularly navigating between questions, is easier for the student. Answers are not submitted to the receiving laptop until the student has entered all of the answers into the handset. Navigation is therefore performed internally for each individual handset as its display can show all the state information that is required for the student to easily navigate and answer.

Vote Reception Display

In some systems, it is important to display the votes as they are received, particularly systems which rely on one-way transmission. This visual feedback to users allows them to re-vote as necessary. Superficially, this feature could be considered obsolete where two-way transmission would seem to guarantee that every vote is received but most systems have some display of, at least, how many handsets have responded. The consideration of the human aspect of the system as a whole suggests why: there a too many ways that a button could not be pressed at all (for instance, a student’s attention wandering; mis-pressing the buttons on the handset; etc) and plenty of situations require all present to respond (for instance, summative testing).

Graphing

The summation of a group vote into some quickly understandable form is a feature of all systems, though in some systems this plays a secondary role (i.e. the system is primarily for collection of data for later analysis). The most common form of summary of the data is a bar graph though other representations, particularly pie charts, are also used. Interpreting the data correctly in these forms (and under lecture conditions) takes skill. It is easy to misinterpret the data. For instance, with a bar graph, one third of the class can be interpreted as a majority due to the rest being evenly spread through the other possible answers.

Management and Storage

The data collected is usually managed, to some extent, by the collection software. That is, the data can be manipulated, annotated, deleted or modified in some way within the collection application. The advantages of this are the ease with which data can be found and modified, especially since the application will be aware of the meaning and structure of the data. External applications (Excel, SPSS, etc) may allow more flexibility in manipulating the data, but as a result, they also allow inappropriate modifications to be performed: does the average button value of an MCQ have any meaning?

The first disadvantage of using the collection application to manage data is the necessary attachment of data to a particular machine. Any data collected on one machine (for instance, a machine permanently installed in a lecture theatre) must be moved manually to any other machine on which the data is to be used (for instance the lecturers own machine). Another problem exists with the realtime use of the data: data is often held internally by the collection application and only becomes available at the end of a session or is held in a secured database, inaccessible to any other applications. Real-time uses of the data are therefore limited to what can be achieved by interrogating the data store, be it internal or external.

Systems differ in what information is attached to data at the point of collection. Annotation of the data provides some context for later interpretation. Some systems record only a bare minimum whilst others seek to capture as much information as possible.

Reporting

Most systems allow the collected data to be analysed after collection. Generally, the expectation is that this will occur after the lecture and is along the lines of automatic marking. Some systems allow the data to be summarised in different ways: by student; by question; by session; by class etc. These reports can be exported to allow editing outside the system, for instance in a spreadsheet or text editor. Users can develop a fuller understanding of their questions and use of the system. Where a student can be identified across sessions, these automatic reports can easily show each student’s progress (and attendance). In systems, which attach no question and/or answer text to the data, this must be post hoc attached, particularly where questions have right and wrong answers and automatic marking is used. Most systems provide a mechanism for extracting raw data to allow more flexible analysis, but the format for this must be interrogated and understood before it can be used.

2.3  Existing Hardware & Software

There are many manufacturers of EVS and most provide a range of systems, each with different capabilities. The previous sections looked in general at the characteristics of EVS, but it is worth looking in more detail at the various systems that are available. To begin, the IR hardware of the ‘Personal Response System’ (PRS) is investigated in detail since this hardware has formed the backbone of the work presented in later chapters. Following this, other currently available systems, both hardware and software, are outlined. This includes an RF version from the PRS manufacturer and their software for use with both IR and RF version handsets.

Personal Response System

Whilst other system have been tested and considered, the Personal Response System (PRS) IR hardware is used here as the exemplar hardware for developing voting handset use in Higher Education. For this reason, it is worth looking in some detail at this particular hardware.

The PRS hardware was originally developed by EduCue but was bought and is now distributed by GTCOCalComp as InterWrite PRS (GTCO Calcomp, 2006). It consists of many small handheld transmitters, illustrated in Figure 2.1, and one or more receivers, shown in Figure 2.2, which are connected directly to a standard serial port on a PC, often a laptop.

Transmitter Handset

Figure 2.1: A PRS Handset.

The handsets have an on/off button, ten buttons, representing the digits 0 to 9, and two extra buttons that allow confidence levels to be set. There is also a single LED capable of displaying green, red or yellow representing the current confidence level.

When it is switched on, the handset displays a green light. This indicates both that the handset is on and that the ‘normal’ confidence level is currently set. The digits 0 to 9 are transmitted by simply pressing the corresponding button. The LED flashes for a short period and then returns to solid green. In order to transmit a digit with a confidence level other than ‘normal’, the confidence button (‘H’ for high confidence; ‘L’ for low confidence) is first pressed followed by the digit button. As the system is infrared and only one way (from the transmitter to the receiver), the data can be lost or badly received. This may be due to clashing votes or simply to misalignment. The system is designed around this possible loss of data and therefore, requires that each handset ID be displayed as it is received, on a publicly visible screen.

Each handset is coded with an ID number from 0 to 999999999 (i.e. a billion). Two types of handsets are available: the first type allow the code number to be reprogrammed whilst the second is known as ‘Fixed ID’ and does not allow reprogramming.

The transmitter is fairly high powered and works up to a range of 99m in a 45 degree cone. The handset is powered by two AAA batteries in a compartment in the back, which can be secured with a small screw.

Receivers

Figure 2.2: A PRS Receiver.

The receivers have a small aperture and a red LED. They have two standard 1/8” jack sockets, one marked ‘IN’, and the other marked ‘OUT’.

If more than one receiver is to be used, they can be daisy-chained together by connecting the ‘OUT’ socket of one to the ‘IN’ of the previous receiver and so on until the first receiver. The first receiver has its ‘OUT’ socket connected to the serial connection cable described in the section below. The manufacturer recommends around 30 to 50 handsets per receiver so in a class of, say, 300 at least six receivers would be used.

When the power is connected, the LED on each receiver blinks several times and then remains lit indicating that it is connected. When a handset transmits to a receiver, that receiver’s LED will blink once.

Unfortunately, if the chain is improperly connected and one of the receivers has its connections reversed then the lights in all receivers will blink and light but the incorrectly connected receiver (and all subsequent ones) will not be able to receive data. Therefore, there is no immediate way of knowing whether the receivers are properly connected (other than if no data is received)

Serial Port

The receiver hardware needs to be connected to a standard PC serial port. The ‘OUT’ of the first receiver is connected to the serial cable which has a standard DB9 serial connector on one end and splits into a power connector and a standard 1/8” jack socket at the other end.

Newer laptops often do not have the required serial connector but do provide USB sockets. A serial-to-USB converter can be used on these newer laptops. This is also the method used with Apple laptops, which are now supported by the latest version of the manufacturers software.

Other PRS Hardware & Software

As mentioned above, the PRS IR handsets were originally developed by EduCue. This was a commercial venture based on work done at Hong Kong University of Science and Technology by Nelson Cue (CELT, 2006).

EduCue PRS Software

The original software for the PRS handsets was a Microsoft Windows application, that provided minimal functionality, including: receiving, decoding, summarising and displaying the vote data; the ability to load class lists; and some automatic marking. The software was limited to MCQs only but this could be enhanced by the use of the PRS confidence levels. Many problems existed with the software but it was extremely reliable and most problems were of small moment.

InterWrite PRS RF Hardware

The PRS RF handsets have ten digit buttons, plus/minus and decimal point buttons, five letter buttons (A-E), True and False buttons, a ‘send’ button, two navigation buttons, a delete button and a button to access an internal menu system for control of the more advanced features. These advanced features include a ‘homework’ mode, which is a disconnected use of the handset to enter the answers to up to 99 questions that can be collected later during a regular voting lecture. The handset has a two-line alphanumeric display, which is used to display class names, entered answers and the transmission and reception state of vote data. Answers can be single alphanumeric digits, sequences of digits, numeric values (including negatives, decimals and fractions) or short text answers. The battery life is around 12 months. The RF receiver connects to a USB port but requires its own power supply. Each receiver is capable of handling over 2000 handsets.

InterWrite PRS Software

The InterWrite software to support both IR and RF handsets includes four main modes of use, however only one of these directly relates to data collection. The others are data management tools relating to class set up, marking and the preparation of content and questions for display at a later point. An add-in is available that allows the collection of data to be tied to a PowerPoint slideshow. An alternative to this allows a screenshot of the computer display to be automatically taken at the point that data collection occurs, in order to provide some record of the session (beyond simply the data collected). Little analysis of the data can be performed during a live session: essentially questions are distinct, immutable units. Reporting is available to summarise by session, by question or by student but only after the fact.

ClassTalk

This system is now unsupported but is presented for historical reasons. It was based on Texas Instruments TI calculators or a Hewlett-Packard PDA and, unlike the other systems presented here, required a wired connection between them (at least, between the handset and the ‘base station’ which may use RF communication to connect to the receiving computer). The software was developed by Better Education (Classtalk, 2000) however they are now working directly with Texas Instruments on a system called TI-Navigator (Texas Instruments, 2006). These systems are of interest in that they rely on a general purpose device (i.e. a calculator) rather than bespoke hardware devices.

The ClassTalk system was capable of simple MCQs and questions requiring multi-digit responses. Some text could be downloaded to be displayed on the student handset as a prompt or reminder of the question and answers. These capabilities were well beyond those of its peer IR or RF systems of the time. The system was essentially designed for small classroom use. The limitation of requiring a wired or semi-wired connection impacts heavily on the feasibility of its use in Higher Education, especially in large classes: it is not easy without permanent installation of the equipment and this necessarily ties it to one location. The TI-Navigator system is more advanced than ClassTalk and provides many features that lift it beyond the realm of an EVS. Indeed, the terminology used for these two particular systems is a Classroom Communication System. This name correctly suggests that the system can be used as the medium via which communication in the classroom occurs. This can take the form of MCQs, but additionally text can be sent and received, the operation of the calculator can be publicly displayed and applications can be shared.

Classroom Performance System

The Classroom Performance System (CPS) handsets are made by e-Instruction. They currently have two different systems, one based on IR and the other on RF connection. The information required to decode data from the handsets can be found on the e-Instruction website (e-Instruction, 2006). This means these handsets can potentially be used with a custom software system. Their own software is fairly extensive, providing methods of preparation and presentation of content and extensive methods of reporting on collected data. Both handset systems allow standard MCQs, multi-digit, multi-answer or multi-question use.

CPS IR

The CPS IR handsets have five buttons representing answer options and three navigational buttons that can be used in a self-paced mode. They have no visual feedback that an answer has been sent and since they use one-way transmission (from the handset to the receiver), they require the user to revote until their handset identifier appears on screen.

By default, two hundred and fifty six handsets can be used in one class though the e-Instruction website suggests a Higher Education model is available which will support an unlimited number of handsets. The receiver unit is connected to a standard serial port and requires its own power supply. Suggested battery life for the handsets is 1-2 years.

CPS RF

The CPS RF handsets is available in several versions: K-12 (North American primary and secondary education), Higher Education and now a ‘2nd Generation’ handset.

The K-12 model allows 300 handsets per receiver. The handset has the full ten numeric digit buttons along with buttons for plus/minus, decimal point and various navigation buttons. This handset also has a two-line alphanumeric LCD display.

The HE model allows 1000 handsets per receiver and has a similar set of buttons. It does not have an LCD display but instead has two indicator lights.

The ‘2nd Generation’ model provides all of the features of the other versions and is a smaller physical size. Some extra features are also added: the handset supports equation entry; can provide correct/incorrect feedback; and can be used in a disconnected mode. It has a three line LCD display, the top line showing status icons such as signal strength.

OptiVote

The UK representatives of e-Instruction (e-Instruction UK, 2006) have developed their own software and re-branded the CPS IR equipment as well as releasing their own RF handset. This handset has ten digit buttons, various navigation buttons and an LCD display (Optivote, 2006).

H-ITT

The H-ITT handsets have ten buttons for the ten decimal digits and also provide three navigation buttons. The feedback to the user that a vote has been sent is via a single LED. The batteries can be expected to last 12 months.

The H-ITT website (H-ITT, 2006) states that under testing their IR handset system recorded 200 votes in 10 seconds and attributes this to the fact that the IR communication is two-way between handset and ‘receiver’ base unit. The recommended ratio of students to receiver is 50 students to one receiver however there is effectively no limit on the number of handsets in any one class. The receiver is connected to and powered by a USB port and thus does not require an extra power supply.

The company has recently announced the introduction of RF handsets, which operate even faster than their IR handsets: 1000 responses in 1 second using only one receiver. The handsets have the same buttons as the IR system. An earlier version of the H-ITT handsets was available which only included five answer buttons and a ‘menu’ button but these have been discontinued.

Information about the data that is sent by the handsets is available on the H-ITT website mentioned above. This would allow the incorporation of this hardware into third-party systems such as the one presented in Chapter 4. The handsets can be used with either single MCQs or, in testing mode, with a bank of questions which students navigate through and answer at their own pace. The provision of multi-digit responses or multi-answer responses seems to be absent.

Qwizdom

Qwizdom have produced three versions of their handsets: Q3, Q4 and Q5. The company has abandoned the use of IR in favour of RF technology and their website (Qwizdom, 2006) only advertises the Q4 and Q5. The Qwizdom software includes a comprehensive suite of preparation and presentation tools and a separate application can be purchased which integrates with Microsoft PowerPoint

Qwizdom Q3

The Qwizdom Q3 handset has now been discontinued. These were a two-way IR based handset with many buttons: the full ten numeric buttons; decimal point, space and fraction buttons; True and False buttons; a ’SEND’ button and 3 navigation buttons. A red and a green LED provided visual feedback to the user confirming that a vote had been sent. This allowed the handset to indicate that a vote had been received, where both LEDs would flash, or that a correct or incorrect answer had been given by lighting the green or red LEDs individually. The maximum number of handsets that can be used in one class is 255. The receiver unit is connected by USB and does not require its own power supply.

Qwizdom Q4 & Q5

The Q4 and Q5 handsets are similar to the Q3 in many respects, including their buttons, except the Q5 has a multidirectional navigation button. The major differences between the Q3 and the Q4 or Q5 are the use of RF technology and the addition of an LCD display. This allows the handset to be used with questions that require single digits, multiple digits (including textual answers), multiple answers or where multiple questions are to be answered. The receiver is again connected to a computer through a USB port. There is effectively no limit on the number of handsets that can be used in the one class.

IML

The IML handset system is the Rolls Royce of handsets. Full details of the system and its capabilities can be found on the manufacturer’s website (IML, 2006). Some of these capabilities include text entry, a built in microphone (which can broadcast sound back to the receiver), a built in earpiece (which can receive broadcast sounds), a smart card system to easily associate a user with a handset, rechargeable batteries and an anti-theft alarm. The handset also has a large LCD display that is capable of displaying more than simple characters. It is unclear how extensive the display capabilities are and how easily this could be modified.

The software is similarly feature endowed. PowerPoint can be used to prepare a lecture with questions of many different formats including simple multiple choice and ranking questions (i.e. “Put these elements in order”). Multiple digit responses, including text, are possible, as is the handsets use in an offline testing mode. Some analysis of the data can be performed in live situations including before and after comparisons and crossing of questions (See Sections 3.3.2 and 4.3.2).

TeamWorker

The TeamWorker handsets comprise ten digit buttons (though only 9 answer options are allowed for any one question) and a two-line LCD display. The website (Teamworker, 2006) suggests that there are essentially two main modes of use: ‘Group Work’ and ‘Survey’. These are essentially the connected and disconnected modes, respectively, as discussed in Section 2.1.1. In a connected mode, 512 handsets can be used at once whilst in a disconnected mode, only 32 handsets can be used. The description of the system seems to suggest only single digit answers are allowed for any one question. The transmission method is two-way RF and the receiver is connected to the computer through a USB port.

EduClick

EduClick is an IR-based system though the details of this system presented on the manufacturer website (Aviro, 2006) are limited. The handsets would appear to have the full compliment of ten digit buttons and, potentially, some navigation buttons though the documentation is unclear how these could be used. The receiver is connected to a serial port and requires its own power supply.

The software supports five question formats, which include standard MCQ and ‘fastest response’ questions. Documentation suggests seven report types can be produced including by student or by group but the exact nature of these is unclear. In preparing questions, the use of images and video are allowed but support for PowerPoint is limited.

Interactive Presenter

Interactive Presenter produce handsets that are credit card sized and come as either RF or IR models. Both models use two-way transmission and the receivers are connected and powered by a computer’s USB port though the IR version requires an extra power supply if multiple receivers are needed. The Interactive Presenter website (Dolphin Interactive OY, 2006) suggest the IR version is capable of use with audiences of up to 200 and the RF version for up to 4000. The handsets themselves have the ten decimal digits and feedback on the transmission and reception of data from the handset is provided by coloured lights, either from a single LED in the case of the IR system, or from three separate LEDs in the RF system.

The software provides many formats for question preparation and presentation, including the use of PowerPoint. Analysis of collected data can be performed including question crossing and splitting as described earlier in Section 2.3.11.

TurningPoint

This manufacturer sells a bespoke software system along with the ResponseCard hardware. The software can be used with other hardware systems however the details of exactly which systems are not stated on the website. Perhaps the most interesting development of this manufacturers system is the vPad virtual handset, which runs on a Microsoft Windows desktop PC, laptop or handheld device. The vPad is discussed below, however full details of the TurningPoint software can be found on the website of Turning Technologies (Turning Technologies, 2006). TurningPoint is a question authoring tool for PowerPoint but allows the handset technology to be used in a variety of ways. Standard MCQs are available but the technology can also be used in a number of other ways.

The first of these is to use the handsets in a competitive manner. The allocation of handsets to teams can be achieved allowing data to be collected and summarised by team scores, individual scores or by ranking by the speed of response to the most recent question.

Another use that the software provides is ranking. Participants can select more than one option from a set and responses are combined using a weighting based on the order they are selected.

Some comparison of data from up to 4 question can be set up to be displayed within PowerPoint and a feature called ‘Conditional Branching’ allows the flow of a presentation to be controlled by the responses given to a particular question. For instance, the question ‘Should we talk about A or B?’ can be posed and dependent on the audience response, different subsections of the PowerPoint presentation would be subsequently shown. The software also defines a special question type called ‘Demographic’ which can be used to split the results of subsequent questions by the demographic categories found in the first question. Various report types can be automatically produced including summarising data by the questions or by the students.

The ResponseCard hardware, described below, is currently sold with the TurningPoint software, although the software does operate with other hardware as mentioned above.

ResponseCard IR

The IR version of the ResponseCard has ten digit buttons and a single LED to indicate transmission of a vote. As this handset is based on one-way IR transmission, a visual feedback display is used in the software component (i.e. in TurningPoint). The battery life is approximately 12 months.

One receiver is recommended for 80 students with no maximum number of handsets in any one class. The receiver connects to a computers USB port and draws power from this, therefore not requiring an extra power supply.

ResponseCard XL

The ResponseCard XL is again an IR based handset but has a large LCD display and extra buttons for navigation. The handset can be used in a disconnected mode allowing a bank of questions to be answered before submitting to a receiving computer. There are both audible and visual feedback of transmission and reception of data, either as single answers or as a bank of answers.

The receivers for the IR and XL are similar, connecting to a USB port and requiring no extra power supply.

ResponseCard RF

As the name implies, the ResponseCard RF is based on RF transmission and this is two-way transmission. Feedback of successful transmission and reception of a vote is provided by a single LED. The battery life stated for the RF version is 6 to 12 months, less than that of the two IR versions.

The receiver is again connected and powered from a computer’s USB port but is much smaller than for the IR system. 1000 handsets can be used with one receiver and there is essentially no maximum number of handsets in any one class.

vPad

The vPad is intended to be installed on PCs, laptops or handheld devices and communicates with the receiving computer using standard wired or wireless network technology. All of the features of the ResponseCard hardware are available however, the availability of a keyboard (either real or virtual) means free text input is also possible. One suggested use for this within the software is for the students to submit questions to the lecturer, either to be answered by the lecturer or to be presented back to the rest of the students for them to answer.

Impact Explorer

This software, produced by Banxia (Banxia, 2006), is compatible with several handset systems: PRS IR handsets (describe in Section 2.3.1), the Fleetwood Reply WorldWide handset (describe in Section 2.3.14) and the ResponsePad handsets. The software is aimed primarily at business users and this is perhaps reflected in the key modes of use. These include standard MCQ, a two dimensional rating of items (i.e. cost vs. benefit), and ranking which allows items to be prioritised. Demographic splitting of question data can also be performed. There is also a method of integration with PowerPoint.

ResponsePad

The ResponsePad handsets are two-way IR based transmitters with ten digit buttons and three special function buttons. They are the default handsets sold with the Impact Explorer software, described in Section 2.3.12. The transmission and reception of data is indicated by a single LED on the handset. The receiver is connected to a serial port and requires its own power supply. A single receiver is used and can support up to 512 handsets.

RxShow

The RxShow system is a software-only system for using voting technology within PowerPoint. The manufacturer’s website (Socratec, 2006) suggests a range of handset hardware can be used in conjunction with RxShow, including PRS, HITT and the Fleetwood ranges discussed earlier. Up to 250 RF handsets and 350 IR handsets can be used at once. Many question formats are supported, including some ‘game’ modes. The graphing of data is drawn directly onto PowerPoint slides and this can be saved after a lecture to provide a record of the session.

FleetWood

The Fleetwood website (Fleetwood, 2006) states “Fleetwood invented 2-way wireless audience response technology” and whilst they have built and sold a range of systems, there are currently only two systems available. These are used with third-party software usually designed for a specific task.

Reply IQ

This version of the Fleetwood Reply system includes a large display allowing up to six lines of 20 characters to be displayed. The handset has the ten digit buttons but includes five ‘soft-keys’. The function of these ‘soft-keys’ can be changed and a label for their function can be shown on the handset display. The display shows both the user entered data, which can be numeric or alphabetic, but can also display messages sent from the receiving computer. This would allow the question and answer options to be displayed on the handset itself rather than on a publicly visible screen. The system requires one receiver per 1500 handsets and the maximum number of handsets that can be used at once is well over 20000. The connection to a computer can be either, directly by a standard serial port, or through an Ethernet network.

Reply (Worldwide)

This is the smaller version of the two FleetWood systems with 15 buttons (the full ten digit buttons and 5 ‘soft-keys’) and a single digit display, which shows the users input and confirms the reception of vote data. Battery life is stated as [this got deleted in conversion to pmwiki] responses rather than as a time period. The ratio of handsets to receivers is 250 handsets to one receiver. The maximum number of handsets is over 3000 but special configurations can be purchased that allow over 18000 handsets to be used in one session. The receivers are connected to the receiving computer by USB.