Welcome to QAnalys! This software was developed by Tommy Wasserman (design) and Hans Alshoff (programming)! QAnalys is freeware. However, if the software is used in a publication, electronically or in print, the source (Tommy Wasserman & Hans Alshoff) should be indicated somewhere in the publication. QAnalys is an add-on extension for Excel. The same file will work on a PC as well as a Macintosh, as long as Microsoft Excel 5.0 (or higher) is installed. Installation of extension Open Excel and choose extensions from the Tools menu. Activate the browse-button and choose the file qanalys.xla and close (the file should now be copied into the macrolibrary of Excel). The extension is now loaded and visible in the bottom of the Tools menu (QAnalys with submenu). Usage Enter your textual data (from all significant variants in a textual apparatus) on a separate sheet (in this example sheet 1). Enter the names (sigla) of the manuscripts in the first row (choose whatever fonts installed). In the first column (to the left) it is appropriate to place the data of the base text. Each row below the MS-names represents a textual variant and thus the base text column should contain the digit "1" for as many variants as are entered. All manuscripts that agree with the base text in a variant are assigned to that same value, i.e. "1". Then another variant is assigned "2", if there is a third variant "3" and so on. If a manuscript has lacuna or a singular reading in a variant it should be assigned value "0" (and it will be excluded from the calculation at that particular point). Some scholars argue that singular readings should be included in the statistical calculation and of course it is possible to assign the singular reading a unique value and thus allow it to be included in the calculation (this is not done in the example below). Example: TR MS1 MS2 MS3 MS4 MS5 (manuscript names in row 1, 1 0 1 2 2 2 Textus Receptus=basetext) 1 2 1 2 3 3 (variant 1 in row 2) 1 1 1 1 1 0 (a singular reading in MS5, 1 2 2 1 0 2 assigned "0") 1 1 1 2 2 1 (lacuna in MS4) When all the variants are entered, in this example only five variants, it is time to conduct a quantitative analysis. Go to the tools meny and choose QAnalys and Quantitative analysis in the submenu. You will now see a dialogue box. Begin from the top in the "Worksheet" textbox and indicate the name of the sheet where you have entered your textual data, in our example we write sheet1. In "Number of witnesses" we write 6 (including the TR basetext). In "Number of variants" we write 5. Now go down to the output-area. Here you indicate on which sheet you would like the output-data, i.e. the statistical table to be placed. We write sheet2. (The clear box is checked by default and this means that target sheet will be cleared from any old data, before the new data is written). On sheet2 you the following table should now appear (the data is given in percent by default): TR MS1 MS2 MS3 MS4 MS5 TR 50 80 40 25 25 MS1 50 75 50 33 66 MS2 80 75 20 25 50 MS3 40 50 20 75 25 MS4 25 33 25 75 66 MS5 25 66 50 25 66 Whenever you change data in the variantlist in sheet1, the table with automatically update (or if your automatic calculation is deactivated, which is done in the tools menu under options, press shift+F9 to update activated sheet). Note that this applies only to variant data that is changed (not if you enter the name and data of a new manuscript, then you must perform the analysis afresh). Additional options and features You can now experiment with the same example in which you can choose in the quantitative analysis to exclude some of the variants (remember that variant 1 is in row2, etc). This is useful if you want to exclude a variant for some or other reason. You may also choose to have the output data in fractions (not percent). Then you will be able to see how many variants the statistic comparison is based upon. This is useful to be able to evaluate the statistical significance, i.e. if a manuscript has a lot of lacuna it will not be evident from the figures given in percent. Another option is to choose only a part of the total number of witnesses (or variants). The software will automatically include witnesses from left to the right, and variants from the top to bottom (but remember that you have a special textbox in which you may exlude whatever variants). Non-majority analysis With this feature you may enter the textual data of the Majority Text. This must be done in the last column (to the right). TR MS1 MS2 MS3 MS4 MS5 MT (Majority-Text in the last 1 0 1 2 2 2 1 column) 1 2 1 2 3 3 2 1 1 1 1 1 0 1 1 2 2 1 0 2 1 1 1 1 2 2 1 1 Then you choose the QAnalys and Quantitative analysis submenu. The dialogue box will appear again and this time you check the box "Non-Majority count". Be sure not to include the Majority-Text in the "Numbers of witnesses" (in this example write 5 again). In the non-Majority analysis it is very useful to get output in fractions, especially when you work with Byzantine witnesses since these will compared in few or any variants (and thus it is important to see if the statistics is significant enough, i.e. based on enough comparisons). In this example (of course with a far too small sample) we choose to place our output data on sheet3 where the following table will be generated: TR MS1 MS2 MS3 MS4 MS5 TR 0/0 1/1 0/0 0/1 0/1 MS1 0/0 1/1 0/0 0/0 1/1 MS2 1/1 1/1 0/0 0/1 1/2 MS3 0/0 0/0 0/0 2/2 1/1 MS4 0/1 1/1 0/1 2/2 2/2 MS5 0/1 1/1 1/2 1/1 2/2 Comparisons in tables This function allows the user to generate the data into a more readable format, namely a table for each MS with the other MSS compared in descending order. The function "Comparisons in tables" is available in the toolsmenu under QAnalys, below the Quantitative analysis. This function will only work when you have generated at least one table where the output is given in percent (fractions cannot be arranged in this way). You may include the sheet where you have performed your non-Majority analysis and this data may be in fractions and will be brought into the corresponding place according to the percentage data. In our example we first redo the ordinary quantitative analysis and include the MT as a witness and choose sheet2 as outputsheet. Then we choose the "Comparisons in tables"-function in which we enter the total number of witnesses (including the MT), then below we write "sheet2" (worksheet with input data) and we choose to include the generated non-Majority comparisons that we have in fractions on sheet3. For output we write "sheet4". The following table will appear on sheet4: TR MS1 MS2 MS3 MS4 MS2 80 1/1 MS2 75 1/1 TR 80 1/1 MS4 75 2/2 MS3 75 2/2 MT 80 MT 75 MS1 75 1/1 MT 60 MS5 66 2/2 MS1 50 0/0 MS5 66 1/1 MT 60 MS1 50 0/0 MS1 33 0/0 MS3 40 0/0 TR 50 0/0 MS5 50 1/2 TR 40 0/0 TR 25 0/1 MS4 25 0/1 MS3 50 0/0 MS4 25 0/1 MS5 25 1/1 MS2 25 0/1 MS5 25 0/1 MS4 33 0/0 MS3 20 0/0 MS2 20 0/0 MT 25 MS5 MT MS1 66 1/1 TR 80 MS4 66 2/2 MS1 75 MS2 50 1/2 MS2 60 TR 25 0/1 MS3 60 MS3 25 1/1 MS4 25 MT 25 MS5 25 In this table you can see that MS2 and the TR are 80% in total agreement and MS3 and MS4 75% in total agreement. However MS3 and MS4 share two non-Majority variants. As is apparent from the MS5 table the software will arrange according to percent but here we have the same percentage for MS1 and MS4 (66%) and ideally then it should arrange according to next column so that MS4 is displayed on the top because it has 2/2 non-Majority agreement. This is something for the future development of this software and in fact the source code is available to anyone interested. (Send the designer an e-mail: wasserman@bredband.net). Error messages If you get an error message you have probably entered an unexpected variable in a dialouge box (e.g. "4" instead of "sheet4", etc). If you have questions send a note to the above e-mailadress. I cannot not promise to answer, but keep on knocking on the door :-) Tommy Wasserman, September 2001 Örebro, Sweden