Data Capture using Optical Mark Readers (OMR) or Image Scanning Methodologies (ICR/OCR)

Users intending to automate data capture of their survey information have often made enquiries about Imaging products and the advantages/disadvantages of the same when compared to Optical Mark Reading (OMR). Both technologies can be used to gather large amounts of data quickly and easily without the need for data input through keyboards which gives the added bonus of not only reducing the workforce involved but also maintain a far higher degree of data entry accuracy than can be achieved through manual entry. Another factor to consider is that because of the wide introduction of OMR technology into schools, (a system in every secondary school as well as a large number of primary schools) plus the advent of the National Lottery, which uses OMR technology, most of the survey prospects on the street are familiar with the marking techniques used.

A further drawback to potential users of either OMR or Image technology has always been the need to produce pre-printed survey questionnaires which some organisations have found to be inflexible, costly and incapable of meeting design change requirements at short notice. This situation has now been resolved with the introduction of Windows based software systems which enables users to design and print their own forms at a relatively low-cost and which can meet short notice design changes quickly and efficiently.

Image (ICR/OCR) vs. Optical Mark Reading (OMR)

When comparing the two types of technology (Image/OMR) then potential users should look carefully at the task they wish to undertake before selecting their proposed system. Put simply, anyone intending to gather survey data or undertake Multiple Choice Questionnaires where no complex image capture other than marks on a paper are being collected should then consider OMR which can prove to be both faster and more cost effective. This is highlighted by announcements from all the major OMR manufacturers regarding the introduction of relatively low cost automatic feed machines which are still capable of much higher throughput speeds than a comparably priced Image scanner.

However, should users wish to undertake the capture of other data types such as dates, hand-written numbers or text (names, addresses etc.), then pure OMR cannot undertake this task type easily (if at all) whereas Image scanners can, (still within certain constraints even though the recognition software is improving all the time) but at slower speeds. The main requirement and advantage of Imaging as seen by users is the ability to undertake both OMR and OCR data capture using one machine (including capturing the same type of marks used by OMR’s), but user beware, as there is a downside, and the task should be carefully assessed first before deciding to opt for what may be an empty hope and therefore an unfulfilled dream.

Parallel Stream - Computer Aided Interviewing - University of York

Notes -

Scanner Prices for Image and OMR for a throughput of say 2000 sheets per hour are approximately the same. However, OMR technology allows the system to process the forms interactively (‘on the fly’) at the speed of the scanner. Image technology has to read the image into the system and then carry out a verification process (batch processing) where the user is prompted to make corrections to the form for indistinct or unclear text and marks. It is stated by most Image software vendors that, due to the disparity encountered in different types of hand writing, the speed increase of data entry using Image technology is about four to five times that of manual entry systems due to an accuracy rating of about 85% whereas OMR is at least twenty times faster with a 99.9% accuracy rate but OMR can only read marks and not text.