Introduction to Experimental Linguistics

7Introduction to Quantitative Data Processing and Analysis

This chapter presents the basic principles for the analysis of quantitative data. We start by describing how raw data is generally organized after the data collection. We then show that data follows a certain distribution, and we present one distribution in particular: the normal distribution. We also discuss the different ways to visualize and describe data. The second part of the chapter deals with data modeling for statistical tests, before describing the logic underlying such tests. Then, we briefly discuss some tests that have traditionally been applied to linguistic experimental data, and point out the inherent limitations in these. We see that nowadays there are more reliable models for analyzing this particular type of data, through the use of mixed linear models, for example. We present these models, as well as the results obtained with such analyses. We end the chapter by discussing some questions that may arise while analyzing data.

7.1. Preliminary observations

In Chapter 2, we saw that there are different types of variables which can be measured along different scales. The dependent variables used in linguistic experiments are generally measured along continuous scales, such as reaction or reading times, the number of items retained after reading a text, or an item’s acceptability on a scale from 1 to 10. Independent variables are often categorical in order to compare the data collected in the conditions ...