Homework 8

More on Analyzing & Graphing the Results

This page contains some tips on creating graphs for the results from the Speech Perception experiments.

Preparing the Data

The page on Analysis of the Results provides guidelines for sorting the data files, such that all responses to the same sound are grouped together. Once you have sorted the data, you will probably want to replace the "f" and "j" responses with numerical values (e.g. "0" and "1"). Use Edit > Replace in Excel.

The Analysis page also contains information on creating averages of all of the responses to the same sound. The screen-shot below shows how your data file may look once you have calculated average reaction times and proportions of 'd' judgments in the DT Identification task.

In order to prepare your results for graphing, you'll want to group all of the average judgments and reaction times together in a single block of cells as follows.

In order to create this, first type the names of the categories (e.g. 00dt, 05dt, 10dt etc.) into a new column of cells. The start to paste your averages into the adjacent columns. Use the Edit > Copy and Edit > Paste commands.

However, if you have computed the averages in the way suggested on the Analysis page (that method saves you a lot of time!), you'll find that you can't just use the normal Paste command. If you do, you'll get a result something like the following. This is because by default Excel copies the formula that you used, rather than the numbers that resulted from that formula.

In order to paste the actual numbers, rather than the formulas, you need to first use Edit > Copy to copy the cells that you're interested in. Next, place the insertion point in the cells that you want to paste into, and then select Edit > Paste Special....

This will bring up the Paste Special dialog box. Select Values, if you want to paste the numerical values in your cell, rather than the formula in the cell.

Once you click OK, you should see something like the following, which is what you want.

You can now use the same procedure to paste all of your averages into the same set of cells, yielding something like the next picture below.

Generating the Graphs

If you want to plot "Category" against "Percentage 'd' Responses", then you'll need to highlight two columns of data. If the columns are adjacent to one another, this is trivial . If the columns are not adjacent, as in the picture below, then you need to hold down the Control key (PC) or the Command key (Mac) while selecting the two ranges of cells.

Once you have selected the data that you want to graph, you can click on the Chart Wizard button in the top of your window.

This will bring up a window of choices like the following. If you want a line graph, then simply select Line in the lefthand menu of Chart Types. You can always click on "Press and Hold to View Sample" if you want to see how your chart will look.

You can click Next to follow a series of additional dialog boxes, which allow you to specify other chart properties. Alternatively, if you immediately click Finish, you will generate a basic graph, like the following.

The graph above is a sample graph of judgments for DT Identification. The graph below is a sample graph of reaction times for DT Identification.

A Couple of Notes on Discrimination

1. The data files for the discrimination experiments have an additional column that you don't need. You can simply select the "mouse_down" column and delete it (Edit menu).

2. In order to graph accuracy for the discrimination experiments, you need to convert same/different judgments into correct/incorrect scores. You can use the table on the Analysis page to figure out when a same response is correct, and when a same response is incorrect.

When you convert the "f" and "j" responses to correct/incorrect responses, you'll probably want to treat Item1-Item8 separately from Item9-Item18, since Item1-8 were all the same, and Item9-18 were all different.

Use the standard Edit > Replace command to edit these cells. If you're doing this by hand, one cell at a time, then you're taking much more time than you need to.