You are on page 1of 4

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/307575487

Peer Pressure Questionnaire-Revised

Research · September 2016


DOI: 10.13140/RG.2.2.10861.79842

CITATION READS

1 130,671

1 author:

Sunil Saini
Indian Journal of Positive Psychology
16 PUBLICATIONS 428 CITATIONS

SEE PROFILE

All content following this page was uploaded by Sunil Saini on 03 October 2022.

The user has requested enhancement of the downloaded file.


Sometimes I have to under go peer RELIABILITY:
14. pressure to be liked in a group. 2.78 1.22 10.66** .45** Reliability is one of the most important characteristics of a tool which indicates how
accurately a scale measure whatever it measure. The peer pressure scale is a reliable scale.
There are two basic concerns with respect to reliability, consistency of items within a
Many times I put off my homework
measure and stability of the measure over time. Although reliability may be calculated in a
15. and other important assignments to 2.62 1.34 10.17** .50**
number of ways, the most commonly accepted measure is internal consistency reliability
for friends' party.
using Cronbach's Alpha (Price & Mueller, 1986).
Sometimes I have to appease my peers
Internal consistency:
16. by doing things that I don't want to 1.22 10.42** .48** Internal consistency refers to the homogeneity of the items in the measure or the
2.75
do. extent to which item responses correlate with the total test score. Cronbach's Alpha assess
To maintain a status in a peer group, whether all the items in a scale are measuring the same thing. The internal consistency of
sometimes I pressurize my parents to the scale was established by using Cronbach's alpha coefficient and a reliability of 0.79 was
17. 2.29 1.29 12.44** .50**
buy an expensive item. obtained for final 25 items. Nunnally (1976) suggested that an alpha of 0.70 should be the
minimum acceptable standard for demonstrating internal consistency. Therefore, the scale
I do not take advice from my parents has a good reliability coefficient.
18. 2.30 1.30 9.51** .36**
about peer group activities.
Test-retest reliability:
It is difficult to think about the Test-retest reliability is the index of a measure's stability over time obtained by
19. negative consequences of what we do 1.22 5.14** .30** correlating the results from two occasions of assessment. For the analysis of test-retest
2.86
with peers. reliability (temporal stability coefficient), 25-items scale was used on the same sample after
a period of 60 days. The procedure for the retest session was identical to that used for the
There is no harm in doing one wrong initial data collection. The coefficient of temporal stability was measured by using Pearson
20. with friends when we do a number of 2.30 1.28 8.72** .38** product-moment correlation method and internal consistency of the scores was measured
good things with them. by using Cronbach's alpha. The results obtained indicated high test-retest reliability
(r=0.33**, p<.01) and internal consistency (á= 0.77**).
It is very difficult for me to deny
21. friend's request to drink in a party or 2.86 1.26 8.29** .31**
VALIDITY:
in other occasions.
A tool is said to be valid if it measure what it claims to measure. The validity of a
Sometimes I do risky and harmful measure is examined in a number of ways, viz., face validity, content validity, predictive
acts to get acceptance in the peer validity, concurrent validity and construct validity.
22. 2.57 1.28 11.26** .52**
group.
Face validity:
The face validity of an instrument is established when the items in the instrument
When I feel uncomfortable in a group
23. 3.25 1.26 7.27** ..32** are clearly and obviously related to the phenomenon being measured, when the items are
I do not how to say ‘NO’.
relevant to the stated condition or purpose of the instrument and when the items are based
I usually compromise with peers' upon whatever knowledge is available at the time of construction (Breakwell et al., 1995).
24. request for a movie, party, etc. 2.12 1.30 8.39** .37** The items of the Peer Pressure Scale satisfy these conditions.

Content validity:
At times I feel pressure to watch
25. 2.51 1.36 8.64** .43** Content validity refers to the adequacy with which a measure assesses the domain of
pornography.
interest. Content validity of the items was further established by using expert technique

8 9
(Nunnally, 1978). Initially 62 items were constructed to for the peer pressure scale. Both support the validity of newly developed tests or scales. For the present data analysis, we
positive and negative worded items were included in the initial version. Four subject experts, have used principal component analysis (PCA). The Kaiser-Meyer-Olkin (KMO) technique
senior professors with psychometrics background, evaluated the face and content validity of was used for evaluating factorability. This technique indicates the extent to which a
the scale. These experts were selected on the basis of their expertise in psychological test correlation matrix actually contains factors or simply chance correlations between a small
construction. The experts opined that the scale has good face validity and content validity. subset of variables.

Criterion validity: Principal component analysis:


Criterion-related validity pertains to the relationship between a measure and another Principal component analysis yields one or more composite variables that capture
independent measure. Criterion validity evidence obtains when the measure being validated much of the information originally captured in a large set of items. The components are
correlated with others measures that are designed to assess the related constructs (Kazdin, defined as weighted sum of the original items (DeVellis, 2003). Principal component
1995; Campbell & Fiske, 1959). There are two types of criterion validity: (a) predictive analysis with varimax rotation was then used to create factors structure of the 25 items.
validity, and (b) concurrent validity. For the current test, we assessed both predictive and The varimax (orthogonal) rotation is the most commonly used method of rotation
concurrent validity. (Tabachnick & Fidell, 1983) that have many advantages in providing meaningful solution.
This analysis was used to 'reduce set of observed variables into relatively small numbers of
Predictive validity: components that account for most of the observed variance (Marcoulides & Hershberger,
To determine the predictive validity, the authors conducted a pilot study with Peer 1997). A factor loading with absolute value greater than .30 was considered sufficiently high
Pressure Scale with 25-items, and found highly significant relation with variables like HIV to assume a strong relationship between a variable and a factor. Therefore, in the current
risk perception, alcohol and drug attitude scale. It was found that peer pressure was study, item with factor loading less than 0.30 were deleted.
significantly correlated with HIV risk perception (r=0.22**) and attitude towards alcohol
and drug abuse (r=0.22**). Both studies were conducted on sample of 200 adolescent boys Extraction of factors:
and girls and the authors reported that coefficient of alpha for the scale was 0.80. Alpha Researchers can use numerous criteria to estimate the number of factors for a given
coefficient tells us about how much variance a group of items had in common. item set. The most widely known approaches were recommended by Kaiser (1958) and
Cattell (1966) on the basis of eigenvalues, which may help determine the importance of a
Concurrent validity: factor and indicate the amount of variance in the entire set of items accounted for by a given
A test has concurrent validity when the test gives an estimate of certain performance. factor. For the current scale, six factors were originally extracted, which accounting for 54%
Concurrent validity of a new test is calculated by finding their correlation with an of variance. With respect to determining the number of factors, only factors with eigen-
established measure. When a new test is validated against previous established test, the values greater than 1 were considered as significant (Rummel, 1970). Kaiser (1958) believed
established test is known as criterion. For the current study, Peer Pressure subscale of Peer that eigenvalues less than 1.0 reflect potentially unstable factors. Further, correlation
Pressure and Conformity scale (Brown et al., 1986) was used to check the concurrent validity coefficient was measured on these six factors. It was found that there was a highly
of the measure. To avoid the contamination of the responses on the new scale, the new scale significant correlation in each factor (Table 2). These high correlations suggest that the six
was administered first than the other measure. There was a significant positive correlation dimensions in the PPS are not independent of each other; therefore, PPS should be used as a
(r= 0.38**, p<.01) between the two measures. Therefore, the finding suggests that peer uni-dimensional scale.
pressure scale is a valid measure for assessing peer pressure among adolescents.

FACTOR ANALYSIS:
Factor analysis is a technique used to identify or confirm a smaller number of factors
or latent constructs from a large number of observed variables (items), thus defining the
underlying dimensions of data, but also reflecting the contribution made by each observed
variable in explaining the dimensions of a measuring instrument (Hair, Anderson, Tatham,
& Black, 1995). There are two main categories of factor analysis: (a) exploratory (principal
component analysis) and confirmatory factor analysis. Although researchers may use factor
analysis for a range of purposes, one of the most prevalent uses of factor analysis is to

10 11
Table 2: Correlation matrix of six factors obtained in factor analysis Table 3: Percentile norms for male, female and total sample

Male Female Total Sample


FACTORS Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6
N 249 342 591
Factor 1 - .38** .36** .49** .31** .36**
Mean 64.80 61.55 64.41
Factor 2 - - .35** .45** .34** .37**
Median 66.00 61.00 65.00
Factor 3 - - - .38** .25** .28**
Factor 4 - - - - .29** .35** Mode 72.00 57.00 60.00

Factor 5 - - - - - .23** Std. Deviation 1.301 1.28 13.21


Factor 6 - - - - - -
Percentiles 5 42.00 41.00 42.00

10 47.00 45.00 47.00


Scree Test:
Cattell (1966) used the relative values of eigenvalues to estimate the correct number 15 52.00 47.00 50.00
of factors to examine during factor analysis-a process known as the scree test. The objective
scree plot is used to visually locate an elbow, which can be defined as a point where the eigen 20 54.00 50.60 53.00
values form a descending liner trend (Bentler & Yuan, 1998). Using the scree plot, a
25 56.00 53.00 55.00
researcher examines the descending values of eigenvalues to locate a break in the size of
eigenvalues, after which the remaining values tend to level off horizontally. In the current 30 59.00 54.00 58.00
study, the examination of scree plot and percentage of variance associated with the factors
and the structure of individual factor has led to the conclusion that only one factor provided 35 60.00 56.00 60.00
the best fit of the data. The percentage of variance associated with this factor was noted to
40 62.00 57.00 61.00
be 0.50 which is sufficient for a valid scale. Thus, PPS is not a multidimensional scale, it is a
uni-dimensional scale. Moreover, the PPS measures in the same way as per the theoretical 45 64.00 59.00 63.00
perspective, i.e., behavior or measures in the similar manner as per the existing measures of
50 66.00 61.00 65.00
the scale. Therefore, the construct validity is also not violated.
55 67.00 63.00 66.60
NORMS:
60 69.00 65.00 68.00
Any test is meaningful when it is interpreted and analyzed with reference to norms.
The peer pressure scale has been normed on adolescent boys and girls. The authors' rationale 65 70.50 67.00 70.00
was that peer pressure in adolescents is likely to put them more vulnerable towards deviant
behaviors. They argued that adolescents are most susceptible towards peer pressure then 70 72.00 69.00 72.00
any other age group. The final norm tables were constructed from adolescent boys and girls
75 72.50 71.00 73.00
within the age range of. Separate norms were developed for boys, girls and total sample.
Individual scoring upto 55 are identified as individuals who experienced low level of 80 74.00 73.00 75.00
peer pressure, individual who score between 56-72 experienced moderate level of peer
pressure, while those who score greater than 72 experienced high level of peer pressure. The 85 77.00 76.00 78.00
current norms developed on this scale are based on the sample of adolescents coming from
90 80.00 78.70 81.00
middle class family irrespective of caste, religion and between 16-19 years. Therefore,
further researchers are advised to develop their own norms on the basis of their population. 95 84.00 82.00 85.00
12 13

View publication stats

You might also like