Data Entry

Not only have large sums of money from government agencies been directed towards the UVNSW program but also many hundreds of man-hours invested by volunteers. The time committed by volunteers incorporates both sub-tidal and land-based activities – the conduct of the actual surveys, pre-planning activities, the communication required to organise members for a particular activity, the discussions held at monthly meetings and the list goes on.

Clearly, investments of this order carry with them the responsibility to safeguard the integrity of the data.  For this reason, it’s essential that members of UVGs spend time and effort both in entering and checking the data and in ensuring it is carefully stored.

Dive leaders should pass the completed data sheets onto the UVG member who is charged with entering the data into the Excel spreadsheet template. The data should be entered into the spreadsheet at the earliest opportunity. We suggest two people enter the data – one reads the information out loud to the other who then types the information into the template. However, there could be occasions where only one member enters the data. A quick glance at a data sheet prior to data entry might be able to answer the question: Are the assessment results very different from what might usually be expected for the site? If the answer is yes to this question then contact should be made with both the dive leader and the buddy pair to clarify the issue.

Whatever the case, proof-read your data!  As we all know, typos often occur. While this might not be too much of a problem when typing a document, particularly with the Spelling command in the Tools menu, there is nothing to alert you to a typo when conducting data entry.  Errors during data entry are common and can consist of: hitting the wrong key (the 0 is next to the 9 at the top of the keyboard – a big difference); placing the value in the wrong cell, and entering two values in one cell if you didn’t hit the Tab or Up and Down arrows properly (e.g. entering a score of 1112 for one cell instead of the 11 in that cell and 12 in the next).  Having entered the data, the last thing most people want to do is to go back through them and double-check all values against the original sheet.  But it has to be done!  Where data sets are large, it is often better to have a different person check the data to the one who entered it. Proofreading requires lots of concentration and should be conducted when you are totally alert.  It is not a job to be done quickly at the end of the day as this is the last bit of direct quality control you have before the data is analysed. (Note – it is sometimes possible to pick up errors during the analysis if the researcher has a good feel for the data and detects an anomaly – however, it shouldn’t have to come to that).

The raw data sheets are in fact a form of plastic. These sheets can easily be recycled for the next survey by simply using an eraser, toothpaste or a solvent like eucalyptus oil to erase the information on the datasheet. Prior to this, it’s critical that the data sheets be photocopied. As a general rule, it is good practice to keep all data, and other information relevant to a project, together and to always replace it in the same location (filing cabinet, box file, etc.) after use. 

For some UVGs it may be feasible to also scan the datasheets. For example, many Multi-Function Devices (MFDs) not only photocopy a datasheet but also are capable of scanning the datasheet as well. Alternatively, a dedicated scanner may be available (if you are lucky it might also have a document feeder that will make the process all the quicker). The scanned document is normally generated as a pdf image. Thus not only will UVGs have a photocopy of the original datasheet but also a digital image. A photocopy should also be sent to the National Marine Science Centre (alternatively a pdf image could be emailed). See the website for contact information. Staff will randomly choose samples of datasheets and compare them to computer entries. We will be looking for example for data gaps, checking latitude and longitude coordinates, examining data for outliers or nonsensical entries. Any shortcomings will be referred back to the coordinator of the appropriate UVG. It is also anticipated that ongoing evaluations of volunteer performance will be undertaken, e.g. through participating in field activities or in land-based refresher sessions.

Rest assured that the data entered by citizen scientists into the Excel template and forwarded to UVNSW will be backed up regularly.