Andijon davlat chet tillari instituti ingliz tili nazariy aspektlari kafedrasi chet tillarni o

Multivariate research data
There are three well known multivariate procedures (see Seliger and Shohami, 1989:222-231) 
1)
Multiple regression
, which is used to examine the relationship and predictive power of independent 
variables. In the case of the relationship between L1 and L2, regression would indicate the prediction of L2 
achievement under the influence of L1. 
2)
Discriminant analysis
indicates which combination of independent variables distinguish between two or 
more categories of the dependent variable. For example, a researcher may want to study which combination 
of variables, L1, motivation, aptitude, etc. can best distinguish between two types of second language 
learners (males/females; learning in formal/informal contexts). 
3)
Factor analysis
. In this case the interrelationships between and among the variables of the data are 
examined in an attempt to find out how many independent dimensions can be identified in the data. Factor 
analysis is a procedure frequently used to validate language tests, for example, to check if the items of the 
Cultural tests really measure the cultural competence. 
Experimental research data
When a control and experimental group are used in experimental research, other procedures are 
commonly used to compare results: The t-test, ANOVA and Chi-square. 
At present , most of the data analysis techniques described in the previous sections can be performed with the 
computer. Some common statistical packages are: Statgraphics and SPSS (Statistical Package for Social 
Science). 
 
LESSON 22 
STAGES OF CLASSROOM INVESTIGATION. DATA COLLECTION, DATA ANALYSIS. 
PROBLEMS IN COLLECTING DATA
In the early days of classroom research observation instruments were used which focused primarily on the 
teacher's behaviour. But as language classroom research has become more deeply involved with issues in 
language learning, a clearer focus on the learners, and on the interaction among learners and teachers, has 
superseded the earlier emphasis on teacher behaviour. One result of this shift in focus has been a decrease in 
the use of observation schedules and an increase in the use of discourse analysis of transcribed data.