So while waiting to figure out if will actual do fieldwork this summer, today was playing a bit with artifact weight and size data from assemblages from 48PA201, 48PA551, the Big Horn Canyon sites, and a couple others that Rachel Reckin and Jack Hofman and I have been recording. Had question about couple of our recorded artifact weights and couldn’t get back in to re-measure right away, so thought would play around with the numbers a bit (rainy day — over 50mm rain in last 30 hours) rather than go outside.

In documenting artifacts the linear measurements are fairly straight forward: position the calipers correctly, hit the data button, and send the reading to the spreadsheet. For weights, don’t currently have portable digital entry options so have to take a reading and type in the values and try not to make any transcription errors. So always wonder if got it right and didn’t add an error (try to double check, but sometimes may not) when looked at the scale and then typed into spreadsheet.

Today was wondering of the values I’d recorded for the five artifacts shown at the top of the page were correct. Since there should be a relationship between artifact size and weight, then perhaps the linear data that we usually record (maximum length, maximun width, and maximum thickness) could be used to give an approximation of weight. Calculated a rough artifact volume (lenght x width x thickness — all to nearest .01 mm) and plotted it against weight (recorde to nearest .01 g). This is what results:

Artifact approximate volume and weight relationship.

Pretty tight relationship. So took the next step and used the linear regression equation to calculate an estimated weight for each artifact in the sample (N = 909) and see how closely the they correspond.

Differences between estimated and actual stone artifact weights.

Fairly close. This will be useful in two ways. First, will include a data checking function into our artifact spreadsheet so that when have done the lenght, width and thickness will calculate an estimated weight to compare to the weight entered into the spreadsheet. If more than 1.01 g difference, will get a “Check” message to be sure have not typed the wrong weight value into the column, if less than will get an “OK”. Should cut down on some of the data entry error.

Modification to GRSLE projectile point data entry spreadsheet to tag weight entries that may need double checking =IF((ABS(AN3-AL3)<1.01),”OK”,”CHECK”). Rainy day project of trying to increase data capture accuracy a bit – done.

Second, sometimes in the backcountry when doing “catch and release” work can’t always get artifact weight (sometimes don’t have the scale and sometimes way to windy or otherwise too hard to take), but now can calulate an esitmate for trying to do things like talk about the total weight of recorded obsidian on a site — or other sorts of “yeah, would be fun to have an rough idea about that, but don’t have the data right now” sorts of questions. No as good as have the actual weight, but still better than nothing.

Rainy day amusement — better, than just waiting it out (weight, wait, don’t tell me a bad pun). Little things can make my day, guess am easily amused.