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

Analysing descriptive research data
Descriptive research is generally analysed by means of descriptive statistics. Some of the most common 
descriptive statistical procedures are: 
frequencies
, 
central tendencies
and 
variabilities
. 
a)
Frequencies
are used to indicate how often a phenomenon occurs 
b)
Central tendency
measures provide information about the average and the typical behaviour of subjects. 
- The 
mean
is the sum of all scores of all subjects in the group divided by the number of subjects. 
- The 
mode
is the score which has been obtained by the largest number of subjects, i.e. the most frequent 
score in the group. 
- The 
median
is the score which divides the group into two parts, so that half of the score are above it and 
half are below it. 
c)
Variability
provides information on the differences or spread of the behaviours. It indicates how 
homogeneous (or heterogeneous) the groups are. The most common variability measure is the 
standard 
deviation: 
the higher the standard deviation, the more heterogeneous a group is. Another measure of 
variability used in statistical analyses is the 
variance
, which is the standard deviation squared. 

Correlation research data

---

49 
Correlation techniques are used to explore existing relations between variables. For example, if we 
have data about the students’ achievement in the EFL class and in the L1 class, we can correlate data and study 
the relationship between achievement in L1 and in the FL class. If a 
positive correlation
is obtained 
it means that there is a close relationship between both variables. A 
negative correlation
would indicate the 
opposite view. 
Correlations are indicated by means of the 
correlation coefficient, 
which ranges from – 1.00, indicating 
perfect negative correlation to 1.00, which indicates perfect positive correlation. The significant level obtained 
is very important since it relates directly to whether the null hypothesis is rejected or not. The conventional 
level of rejecting the null hypothesis is p < .05 or p < .01. When reporting correlations the researcher needs to 
specify all these data: the sample size (n) the correlations were based on, as well as the level of significance 
(p).