[Abstract]
[PDF]
[HTML]
[Linked
References]
Statistical Analysis for Assessing Knowledge and Attitude on
HIV/AIDS
A.V. Wadagale1,
P.R. Gangwal2, A.R. Aradwad3, V.A. Jadhav4
1Assistant
Professor, MIMSR Medical College, Latur (MS), INDIA
2Lecturer,
G.H. Raisoni College of Arts, Commerce and Science, Wagholi, Pune
(MS), INDIA
3Assistant
Professor, MIMSR Medical College, Latur (MS), INDIA
4Reader,
Science College, Nanded (MS), INDIA
Corresponding Addresses
Email :
[email protected]
,
[email protected]
Research Article
Abstract:
In this research
article, we describe different statistical techniques for accessing
knowledge and attitude towards the disease called HIV/AIDS. We carry
out questionnaires on 200 students of computer science, stream of G.H.
Raisoni College of arts, commerce and science. The statistical
techniques like Pearson�s correlation coefficient, one way ANOVA,
Tukey HSD, Levene�s statistics are used. Results are produced on
statistical software SPSS 17.0 and MS-Excel. Result shows the
relationship between Knowledge and Attitude on HIV/AIDS. At last, we
give some recommendation, challenges and Major issues focus on
HIV/AIDS.
Keywords:
Pearson�s correlation coefficient, post hoc test, Tukey HSD, Levene�s
statistics, HIV/AIDS
Introduction:
India is second
most populated in the world where more than 1 billion people are
living . HIV infected population is nearly 7% among these 1 billion
that if we transfer it in actual number is reaches approximately 3.9
million+. Out of these actual figure > 1% people are found in 6
states i.e. Maharashtra, Tamilnadu, Karnataka, Manipur, Nagaland and
rest of country shares less than one percent infected population from
HIV/AIDS. Based on annual surveillance data report which is collected
from different sources shows that, men are victims in more than 75%
cases while transmission route is through sex i.e. more than 85%
cases. It also shows that there are significant variations among and
within states. There are different mode of transmission of HIV/AIDS
like sexual contact (84.53%), Blood and Blood products (3.37%), IDUs
(3.36%), Perinatal transmission (2.14%) and other sources (6.70%).
There is need of
knowledge of awareness of HIV/AIDS. For this there is need of
continuous surveillance, awareness program, increased health care
allocations, identification of high risk groups, access to treatment
for all, removal of stigma and discrimination, developing appropriate
guidelines, etc.
In this research
article we collect the information from 200 undergraduate students
from computer science branch. The question arises here that why we
select the students from this stream, there is thinking in society
that most of the students from computer science may not have detail
knowledge of awareness of HIV/AIDS. Considering the changing scenario
of development in India, computer branch play a vital role in many of
the leading management. Here, we check their inadequate knowledge and
negative attitudes in management of HIV/AIDS patient may prevent. The
application of scientific methods of computer science students
resulting into fragmented care of the infected people with potential
negative impact.
Material and Method:
As the number of
computer science students are nearly 200, so we include all of the
students in our study. To know the required information from the
student we prepare the questionnaire which contains questions on
Demographic information, factual questions to know their knowledge,
opinion questions to assess their attitude towards HIV/AIDS. The
question arises here that why we include the students from branch,
because the main issue behind this is, these students have the
technical knowledge of software�s and information system but are the
know anything about today�s hazardous condition of HIV/AIDS. From this
point we can conclude the approximate knowledge and attitude of
society towards HIV/AIDS. To stratify the student�s knowledge we
frame 27 questions out of which 26 questions have two outcomes i.e.
yes or no, while to know their attitude on HIV/AIDS, we fixed 13
questions with three outcomes i.e. Agree, Undecided, Disagree. One
question is place as open format.
Statistical
techniques like ANOVA post doc test, Pearson�s correlation
coefficient, homogeneity of variance is checked by Levene�s statistics
and for comparison between first, second and third year students we
used Tukey HSD which is useful for multiple comparison. All the result
are carried out by using MS excel and SPSS 17.0 software.
Technical Analysis:
Pearson�s
correlation coefficient:
A correlation is a
number between -1 and +1 that measures the degree of association
between two variables (call them X and Y). A positive value for the
correlation implies a positive association (large values of X tend to
be associated with large values of Y and small values of X tend to be
associated with small values of Y). A negative value for the
correlation implies a negative or inverse association (large values of
X tend to be associated with small values of Y and vice versa).
Correlation is symbolically represented by
rxy
|
Score |
Levene Statistic |
df1 |
df2 |
Sig. |
5.814 |
2 |
197 |
.004 |
Table1: Test of Homogeneity of Variances
ANOVA:
When we want to
compare means of more than two groups or levels of an independent
variable, we use one way ANOVA. It is used to find the significant
relations by assuming equal variance. The procedure of ANOVA involves
the derivation of two different estimates of population variance. Then
statistics is calculated from the ratio of these two estimates where
one is between group variance estimate which is measure of effect of
independent variable and other estimate within group variance which is
error variance itself. The F ratio is ratio of between the groups and
within the groups variance. When hypothesis is rejected i.e. when
significant different is lies, post hoc analysis and other test needs
to be performed to get the results.
Score |
|
Sum of Squares |
Df |
Mean Square |
F |
Sig. |
Between Groups |
19.885 |
2 |
9.942 |
1.260 |
.286 |
Within
Groups |
1553.870 |
197 |
7.888 |
|
|
Total |
1573.755 |
199 |
|
|
|
Table 2: One way Analysis of Variance (ANOVA)
Levene�s statistics
for homogeneity of variance:
Test for
Homogeneity of Variances Levene's test is used to test if k samples
have equal variances which is first invented by Great Scientist Levene
in 1960. The term homogeneity of variance is used when there are equal
variance across samples. Levene�s test is less sensitive that we can
used it for small number of samples also. It is alternative test for
Bartlett test. As in ANOVA we assumed that variance are equal overall,
we check this by using Levene�s Statistics.
The test statistics
is W, which is defined as follow:
Where,
-
W
is the result of the test,
-
k
is the number of different groups to which the samples belong,
-
N
is the total number of samples,
-
Ni
is the number of samples in the ith group,
Yij
is the value of the jth sample from the ith group
is the mean of all
Zij
and
is the mean of Zij
for ith group
We check the
significance of W is tested against F(α,k − 1,N
− k) where F is a quantile of the F test distribution,
with k − 1 and N − k its degrees of freedom, and
α is the chosen level of significance (usually 0.05 or 0.01).
Figure 1 : Boxplot
for class Vs. score
Result and Discussion:
We carried out
results in MS-Excel and SPSS 17.0 for Pearson�s correlation
coefficient, post hoc test, Levene�s statistics, Tukey HSD. Here, as
mention, we took place questionnaires on 200 students from computer
science stream. As it estimated, the students from this stream have
less knowledge of the disease HIV/AIDS, the results shows the same. We
conclude stepwise results as follows:
1.
Karl Pearson�s Correlation coefficient shows that there is weak
relationship between Knowledge and attitude (r=0.009) in computer
science students. This means there is need to aware about the HIV/AIDS
in these students. Furthermore, we check their covariance which is
also not significant because value of correlation coefficient is very
weak (r=0.004)
2.
Homogeneity of Variances: 1) The variances of the group are similar.
2) The group should be independent. Since Homogeneity of Variances
should not be there for conducting ANOVA tests, which is one of the
assumptions of ANOVA, we see that Levene�s test [Table 1] shows that
homogeneity of variance is significant (p<0.05). As such, WE can be
confident that population variances for each group are approximately
not equal.
3.
[Table 2] shows that the F test values along with degrees of freedom
(2,197) are not significance of 0.286. Given that p>0.05, we accept
the null hypothesis and reject the alternative hypothesis that there
is no significance difference in scores from different classes. F(2,197)0.05=1.260,
P>0.05.
4.
Post Hoc analysis involves hunting through data for some significance.
This testing carries risks of type I errors. These test are designed
to protect type I errors, given that all the possible comparisons are
going to be made. Post hoc tests are stricter than planned comparisons
and it is difficult to obtain significance. We used Tukey
test/honestly significant difference (HSD) test.
Using Tukey HSD [Table 3], we can conclude that there is no
significant difference of First, Second and Third year and in their
scores.
5.
[Figure 1] gives the information about the class of first year, second
year and third year with their individual scores.
Score
Tukey HSD |
(I) Class |
(J) Class |
Mean Difference (I-J) |
Std. Error |
Sig. |
95% Confidence Interval |
Lower Bound |
Upper Bound |
1 |
2 |
-.270 |
.456 |
.824 |
-1.35 |
.81 |
3 |
.627 |
.527 |
.461 |
-.62 |
1.87 |
2 |
1 |
.270 |
.456 |
.824 |
-.81 |
1.35 |
3 |
.897 |
.570 |
.259 |
-.45 |
2.24 |
3 |
1 |
-.627 |
.527 |
.461 |
-1.87 |
.62 |
2 |
-.897 |
.570 |
.259 |
-2.24 |
.45 |
Table 3 : Tukey HSD for multiple comparisons
From the results we
recommend some major issues and challenges on awareness of HIV/AIDS
like there should be continuous surveillance, Awareness of program, we
must increase health care allocations, identification of high risk
groups, access to treatment for all, developing appropriate guidelines
for HIV/AIDS and removal of stigma and discrimination of this disease.
There are some challenges like training workshops for different
stakeholders, dissemination of national and international guidelines,
behavioral studies for risk reduction, promotion of access to drugs as
all ARVs are available in the market but not in the national program
on HIV/AIDS control.
The major issues
discussed are illiteracy, gender discrimination, imbalanced
globalization, international collaboration with the local interest and
promotion of human subjects in HIV/AIDS research, etc.
Acknowledgement:
We acknowledge to
Principal, Staff members, student and specially NSS department of
Raisoni arts, commerce and science college, wagholi, Pune(MS), INDIA.
References :
[1] Amira Sidig MD, Alsadig Mahgoub DTPH and Abbashar Hussein MD, �A
study of knowledge, attitude, practice towards HIV/AIDS and prevalence
of HIV/AIDS among tea sellers women in Khartoum State (April 2004 -
May 2005)�, SudaneseJ. Of Public Health, Vol 4,No.1: 214-224,2009.
[2 ]Anthony J. Onwuegbuzie, Nancy L. Leech, �Post Hoc Power: A Concept
Whose Time Has Come�, UNDERSTANDING STATISTICS, 3(4): 201�230, 2004
[3] Arvind Pandey, Dandu C S Reddy et al , �Improved estimates of
India�s burden in 2006� Indian J. Med. Res129., 50-58, Jan-2009
[4] Becker N G,�A method of non-parametric back-projection and its
application to AIDS Data�, STATISTICS IN MEDICINE, VOL. 10, 1527-1542,
1991.
[5] Chase, L. J., & Tucker, R. K.,�Statistical power: Derivation,
development, and data-analytic implications.� The Psychological
Record, 26, 473�486,1976
[6] Chase, L. J.,&Chase, R. B., �A statistical power analysis of
applied psychological research�, Journal of Applied Psychology,
61:234�237, 1976.
[7] Chase, L. J.,&Tucker, R. K.,�A power-analytical examination of
contemporary communication Research.�, Speech Monographs, 42(1):
29�41,1975
[8] Cohen, J. �The statistical power of abnormal social psychological
research: A review.� Journal of Abnormal and Social Psychology, 65,
145�153, 1962
[9] Lal S.S., R.S. Vasan, Sankara Sarma P., Thankapran K.R., �
Knowledge and attitude of college students in kerala towards HIV/AIDS,
sexually transmitted diseases & sexuality�, The national Medical
Journal of India, 2000
[10] Longini I M , Byers R H, Hessol N A and Tan W Y, � Estimating
the stage specific numbers of HIV infection using a Markov model and
Backcalculation �, Statistics in Medicine, Vol.11,1992
[11] Overview of making estimates of HIV/AIDS and its impact in
countries with low- level or concentrated epidemics: The Workbook
Method, UNAIDS/WHO, June 2003
[12] Jean-Philippe Laurenceau, Adele M. Hayes, and Greg C. Feldman,�
Some Methodological and Statistical Issues in the Study of Change
Processes in Psychotherapy.�, Clin Psychol Rev., 27(6): 682�695,July
2007