test construction project paper
TRANSCRIPT
Running head: EXTRAVERSION SURVEY 1
Test for Extraversion Survey (TES)
“I have abided by the Spelman Academic Integrity Policy on this assignment”
Ashley Harrington
Spelman College
Dr. Blankson
4/27/16
EXTRAVERSION 2
Abstract
This project examined one of the big five personality traits known as extroversion.
Extroversion is the extent to which an individual is sociable, aggressive and impulsive. The
Personality Test for Extroversion (PTE) was created to analyze this trait. The survey responses
used were taken from a sample of 25 Spelman College students aged at least 18 years. When
analysis was performed using descriptive statistics, it was determined that 100 % of the
participants were both female and black. The internal consistency value, coefficient alpha, was
used to assess both the validity and the reliability of the TES. The coefficient alpha value was
found to be r s=.92, indicating a high degree reliability and construct validity. Discriminant
validity was found between the extraversion, shyness, and antisocial behavior with r=-.203 and
r=-.044 respectively. Further research will have to be done on this test in order to determine its
psychometric quality.
EXTRAVERSION 3
Test for Extraversion Survey (TES)
For the psychometrics test construction project, a test that was intended to measure the
big five personality trait extraversion was developed. Extraversion is a personality trait that was
derived from the modern personality theory known as the five factor model (FFM). It seeks to
contrast individuals that are talkative, sociable, outgoing and caring at the high end of the scale,
with those who are introverted, with a tendency to be shy, reserved, quiet and withdrawn, at the
low end of the scale (Huffman, 2011; Kim & Chung, 2014; Otonari et al., 2012). Impulsivity is
also a trait that is associated with individuals who score high on extraversion (Tiwari, Singh &
Singh, 2009). People who score high on extraversion may also be described using adjectives
such as energetic and self-assured (King, George & Hebl, 2005) According to Carl Jung, those
who score high on in extraversion are believed to align their energy to the exterior world. These
people are thought to be absorbed in what is occurring in the world around them (Carlyn, 1977;
Harbough, 2010).
This test is needed because there is a paucity of research that has been conducted on
people of color, as most research until present date has focused on Caucasians. Therefore, some
of the findings of previous research may not translate to other demographics. This test will fill a
void caused by a lack of diversity in research.
There are four principle uses of objective personality tests. One such use is in the
selection of personnel by employers. Some organizations may seek to identify those that possess
certain personality traits that may be indicators of predictive success in a particular profession
(Hogan, 2015).
EXTRAVERSION 4
The five factor model personality traits appear to be valid predictors of a range of
variables concerning work performance. For example, both neuroticism and conscientiousness
are considered to be predictors of motivation and performance at work. Other traits such as
extraversion, openness and agreeableness are thought to predict a person’s performance and
abilities in specific professions (Dentrick & Chibnall, 2013; Hogan, 2015). Extroversion has
been found to be predictive of ability in areas like management, sales and nursing, but not in
areas like engineering for example (Barrick & Mount, 1991). The big five personality traits may
come together and interact to influences the actions and circumstances of a person, which in turn
may influence job performance and satisfaction (Zhai, Willis, O’Shea, Zhai & Yang, 2013).
Extroverted people are also more likely to report higher levels of job satisfaction because they
are more likely to engage in social interactions with their coworkers (Kim & Chung, 2013).
There are a variety of tests available to measure the construct of extraversion, one such
test is the NEO Personality inventory or NEO-PI. The NEO-PI is a factor-analytically based test
that is intended to measure personality. It identifies five personality factors known as the “Big
Five” personality traits. These traits include openness, conscientiousness, extraversion,
agreeableness, and neuroticism. The most recent version of the NEO-PI is the NEO-PI-3, of
which there are 2 different versions, each with 240 items. Form S, the most popular version, is a
self-report inventory where the subject rates his or herself. Form R of the NEO-PI-3 is a version
of the test that is used for reporting the results of another person. There are also alternate forms
of the NEO-PI-3 form R based on the gender of the respondent. In addition, there is an
abbreviated version of the NEO-PI-3 known as the NEO-FFI-3 which reduces the number of
items from 240 to 60 (Hogan, 2015).
EXTRAVERSION 5
The NEO-PI-3 differs from the previous version, NEO-PI-R, by altering 37 of the 240
existing items. This was done mostly to simplify the reading sections of the test in order to make
the test more applicable to younger populations. It has five domain scales, each corresponding to
the five personality traits found in the Big Five personality model. Each of the five personality
domains contains six facet scores, resulting in a total of 30 facet scores. Each of these 30 facets
is assessed by eight items, resulting in 240 items (Hogan, 2015).
The Myers-Briggs Type indicator (MBTI) personality inventory is a self-report, factor-
analytically based test that is utilized in the assessment of personality types. First described by
Carl G. Jung, the test consists of four domains: Extraversion-Introversion (E-I), Sensation-
Intuition (S-N), Thinking-Feeling (T-F) and Judgement-Perception (J-P). There are currently
four explicit forms of the MBTI inventory, each of which is used in different situations and
scored accordingly. Form M consists of 93 items and is the most common form of the MBTI
inventory. It is typically graded using computer-scoring. Form M-self scorable also has 93
items and is intended to be scored by either the respondent or a professional. There is also the
MBTI Step II Form Q which has 144 items. This version of the test breaks down each of the
four personality domain into five facets in order to provide the test taker with more detail. The
final version of the of the MBTI is the Step III which has 222 items in total. This version of the
test is tailored solely for use in counseling and coaching frameworks where the test taker is most
likely to meet with a professional face-to-face (Myers-Briggs).
Additional tests that measure the construct of personality include the Eysenck Personality
Questionnaire, which seeks to measure neuroticism, extraversion, psychoticism, and lying, and
the Hogan Personality Inventory (Tiwari, Singh & Singh, 2009). The Hogan personality
inventory is an instrument that is meant to describe the more positive aspects of personality and
EXTRAVERSION 6
how they influence an individual’s interpersonal relationships (Salgado, Moscoso & Alonso,
2013).
The target audience of this of the TES includes black, female, young adults enrolled in a
college or university. This is because a test for extroversion is considered a specific domain test,
which tend to have very clear-cut audiences and uses with few exceptions (Hogan, 2015).
Because the TES is a specific domain test, it will most likely have a limited standardization
sample, or norm group. It is even a possibility that there will be no norm group (Hogan, 2015).
A good sample size would consist of at least 30 participants in order to be able to generalize the
results to the population.
Methods
Participants
The sample obtained from the Spring 2016 Psychometric Instruments class consisted of
25 young adult participants. This sample was taken from a larger population of students
attending Spelman college in Atlanta, Ga. The study consisted entirely of female participants of
African American descent, aged at least 18 years.
After the participants were selected to take part in the study, the respondents took the test
home in order to complete it. The tests were then returned to the researcher for scoring. No
compensation was provided as incentive to participate. A total of 20 surveys were recovered, the
remaining 5 were excluded due to incompletion.
Materials
The materials used for this study include the Test for Extraversion Survey or TES, which
is a pencil-and-paper test that was constructed by the author of this paper, Ashley Harrington.
The purpose of the TES is to collect data concerning one of the five factor model personality
EXTRAVERSION 7
traits, extraversion. The final test included 20 items pulled from the original 60 items that were
found on the draft version of the test.
Items concerning level of extroversion were answered by asking questions such as “I am
quick to get involved in social activities at my new job or school” and “I enjoy being the center
of attention.” The responses were given on a five-point Likert Scale: Strongly disagree,
Disagree, Neutral, Agree Strongly agree for both statements. Verbal responses were then
converted to numerical form with assigned values of 1, 2, 3,4, and 5 respectively. The
directionality of some of the negatively-keyed items such as “I do not like being the center of
attention” and “I am slow to get involved in social activities at my new job or school”, was
reversed in order to maintain the consistency of the item scores Item analysis was conducted in
order to elect the items that correlated most with the total score. A final score was then
constructed based on the results of the item analysis.
The Likert scale, often called the method of summated ratings, is a selected response
format that typically employs a five point, three-point, or nine-point scale for each response. It
was chosen as the response format of this test because it allows a respondent to express their
feelings or attitudes, while maintaining the scoring reliability that is typically absent in
constructed response items. The summation of all the responses provide a comprehensive
response to the target construct (Hogan, 2015).
Multiple choice is also a type of selected response format that is also known as multiple
response or forced response items. This format typically uses at least four or five options for
each item, much like with the Likert scale. This format differs from that of the Likert scale in
EXTRAVERSION 8
that the Likert scale can often be adapted into the graphic rating scale to be used by younger
populations who are unable to read, while multiple choice items cannot (Hogan, 2015).
True/false response items are a special type of multiple choice response format that only
has two options: True or False. This differentiates it from both traditional multiple choice
format and Likert format, both of which typically have at least 3 options to choose from. In
addition, much like with traditional multiple choice formats, this response method cannot usually
be adapted for use with those who are unable to read, unlike the Likert scale (Hogan, 2015).
Other measures include Chasity Jackson’s Survey for Antisocial Behavior, which is a 25
item, multiple choice, pencil-and-paper instrument designed to investigate the incidence of anti-
social behavior in school-aged children and adolescents. High scores on this measure indicate
more antisocial behavior problems within the school system, while low scores indicate fewer
behavioral problems. This measure was used in order to assess the criterion related validity of
the TES.
In addition, Maria Reynolds’s Survey of Shyness was also utilized. The Survey of
shyness is a 20 item, pencil-and-paper instrument that seeks to measure an individual’s level of
shyness. The Survey of shyness makes use of the five-point Likert scale with responses ranging
from Strongly Agree to Strongly Disagree, with assigned values ranging from 5 to 1 respectively.
High scores on the measure indicate that the person exhibits a high level of shyness behavior,
while low scores indicate that an individual displays a lower level of shyness behavior. This
measure was also used to assess the criterion related validity of the TES.
Procedures
EXTRAVERSION 9
The surveys employed in this study were composed of a set of responses from students
attending Spelman College in Atlanta, Ga. Responses were a sample collected from a larger
population of all students, freshmen to seniors, who were enrolled at the college. The class that
was selected to participate in the study was a convenience sample. After the 25 participants were
selected to take part in the study, they were given the survey to be completed by hand and
returned to the researcher. The TES was used to measure the Big Five personality trait of
extraversion. The surveys that were collected at the college were done in March of 2016.
Results
Item analysis was conducted via the use of statistical analysis in order to select which
survey items should be included in the final test. By doing this, it became possible to control the
characteristics of the test (Hogan, 2015). Analysis began with the informal tryout phase in which
seven individuals, all of whom were very similar to the target audience, were asked to provide
constructive criticisms about the survey items and instructions. Any items with vague or
ambiguous wording were then removed, while the best items were preserved. Next, came the
formal tryout stage in which a revised version of the test, now containing 20 items, was
administered to the sample of 25 test-takers. Scores were then collected and analyzed using
SPSS computer software.
Since this is a measure that is designed to assess an individual’s level of extraversion, it is
important that this test be able to differentiate between those that possess this characteristic, and
those that do not. Item discrimination refers to an item’s ability to discern between groups of
participants (Hogan, 2015). In this case, groups are differentiated based on a trait that is internal
to the test. The goal is to create groups based on who has more or less of the characteristic being
investigated. Only the items that were found to discriminate best between extraverts and non-
EXTRAVERSION 10
extraverts were chosen. As a result, “high” and “low” groups were created and defined as the
top and bottom halves of the distribution of total scores. Table 1 in Appendix A contains a table
that illustrates the items that were initially written, as well as the final set of items that were
selected.
The reliability of a test is a reference to how consistent a particular measure or study is.
In order for a study to be reliable, it must generate identical or very comparable results when
replicated. Various methods can be employed in order to assess the reliability of a measure
including: test re-test reliability, alternate forms reliability, internal consistency reliability, and
inter-rater reliability (Hogan, 2015).
Test re-test reliability is a form of reliability in which the coefficient is generated by
delivering an identical test on more than one occasion. The correlation between the initial and
follow-up test is the reliability coefficient. The purpose of test re-test reliability is to
approximate any adjustment in the personal conditions of the participants. However, in some
cases, it does not account for variation in test administration, nor does it address any variation
due to inter-rater reliability (Hogan, 2015). Because the test was only administered on a single
occasion, it would be impossible to determine the test-retest reliability coefficient value for this
survey.
Another form of reliability is alternate forms reliability. This is a method for assessing
reliability which necessitates that there be two analogous forms of the same measure, which are
taken under identical conditions. The reliability coefficient is the correlation between the two
scores obtained on the different forms of the test (Hogan, 2015). As a result of there only being
EXTRAVERSION 11
a single form of the TES that was only administered a single time, it was not possible to obtain
the alternate forms reliability coefficient for this survey.
Internal consistency reliability is another method of assessing the reliability of a measure.
This form of reliability can be appraised in a variety of ways: split-half, Kuder-Richardson, and
coefficient alpha (Hogan, 2015).
Split-half reliability, a form of internal consistency reliability, is obtained when a test is
split into two sections and administered in succession. The two sections are then correlated in
order to generate the split half reliability coefficient (Hogan, 2015). A split-half reliability
coefficient was computed for the TES in order to determine the correlation between the two
halves of the test. Findings indicate that the split half reliability coefficient of TES was .81.
Reliability that falls within the range of .80-.89 is considered to be good, however further
information should be used to supplement the test score (Hogan, 2015). Split half reliability only
provides the reliability of half the test, as a result, adjustments were made in order to assess the
reliability of the entire test. This amendment was made through the use of the Spearman-Brown
correction. Spearman-Brown makes the assumption that all of the items that are added to or
removed from the measure are comparable to all of the other items on the test. Based on the
results of this study, the measure was only adequately reliable at assessing the personality trait of
extraversion in test-takers, r s=.79 , p< .05.Although reliability in the range of .70-.79 is
considered to be satisfactory, the usage of this test must be undertaken with discretion and must
be supplemented with information from other sources (Hogan, 2015).
Kuder-Richardson 20, or KR-20, is another measure of internal consistency reliability. It
provides the sum of all the possible split halves for a measure, however it is only useful for
EXTRAVERSION 12
instruments that are scored dichotomously. Therefore, KR-20 could not be used with the TES
which utilized a multiple choice, Likert scale. Coefficient alpha, also known as Cronbach’s
alpha, is an alternate form of internal consistency reliability that would be much more
appropriate for use with the TES because it can be used with any item that possess a continuous
score, such as a Likert scale (Hogan, 2015). The TES generated a coefficient alpha value of .92,
which is considered to be excellent. No additional information would be required to supplement
this test.
Inter-rater reliability, also known as inter-scorer reliability, refers to variations in test
scores due to differences between who scores the test. In order to obtain inter-rater reliability,
the test must be administered to the sample a single time and scored twice by two different
judges independently of one another. The correlation between these two scores in the inter-rater
reliability coefficient. It is imperative that the judges work separately while rating the
instruments in order to avoid one influencing the other’s scores (Hogan, 2015). Inter rater
reliability was not secured for this test because it was only scored by a single judge.
The validity of a test alludes to the extent to which a test measures the construct or trait
that it is intended to. Much like with reliability, there are many different forms of validity such
as content, criterion-related, and construct validity (Hogan, 2015).
Content validity considers the relationship between a test’s content and some behavior or
trait that the test intends to measure. In order for a test to have good content validity, the test
content and the trait that is being investigated must correlate (Hogan, 2015). Content validity is
appropriate for use with the TES because it is important to ensure that an instrument that seeks to
measure a trait such as extroversion, accomplishes this feat. However, certain very specific
EXTRAVERSION 13
forms of content validity such as instructional validity, are not appropriate for this type of test
because personality is an innate characteristic of an individual, and not something that is entirely
taught (Hogan, 2015).
Another form of validity is criterion-related validity. This type of validity establishes a
relationship between an individual’s performance on the measure and some other criterion that is
considered to be important to the construct that is being measured. Criterion-related validity can
take the form of predictive validity or concurrent validity. A test with predictive validity is one
that seeks to measure criterion that will be obtained sometime in the future. In contrast, a test
with concurrent validity would be one that measures the current status of a variable, the TES
would be an example of a test with this form of criterion related validity (Hogan, 2015).
When discussing criterion related validity, it is important to keep convergent and
discriminant validity in mind. A test that displays a high degree of convergent validity is one
that has a relatively high correlation between the measure and the criterion that is assumed to
measure the same construct as that measure. The opposite is true for a test with discriminant
validity. Discriminant validity means that the test and the criterion are not very strongly
correlated with one another (Hogan, 2015). A Pearson correlation was computed in order to
assess the criterion related validity of the TES using the Shyness Survey and the Test for
Antisocial Behavior. It was found that the Test for Extraversion Survey and the Test for
Antisocial Behavior were negatively correlated (r=-.044). There was also a negative correlation
between the Test for Extraversion and the Shyness Survey (r=-.203). These findings indicate
that there is discriminant validity between extraversion, antisocial behavior, and Shyness.
EXTRAVERSION 14
The final form of validity that will be discussed in this paper is construct validity.
Construct validity is demonstrated when a test is found to measure its intended construct. A high
degree of internal consistency is usually an indicator of a high degree of content validity (Hogan,
2015). In order to assess the construct validity of the Test for extraversion, SPSS was used to
conduct a factor analysis. Internal consistency was indicated by a coefficient alpha value of .92,
indicating that there is a possibility that this test would also have a high degree of validity.
However, a high internal consistency is only an indication that a construct is being measured,
other information would be needed in order to determine if the correct construct is being
measured (Hogan, 2015).
Discussion
The final version of the Test for Extraversion Survey or TES, consisted of 20 item that
were designed to measure one of the Five Factor Analysis model traits, extroversion. Responses
were given on the Likert scale with feedback ranging from Strongly Disagree to Strongly Agree.
Verbal responses were converted to numerical form with assigned values distributed from 1 to 5
respectively. Negatively keyed items such as “I find it difficult to open up to new people” were
reverse scored in order to maintain the consistency of the item scores. Scores were entered into
SPSS for analysis and a total scores were generated for each test taker. The high and low groups
for extraversion scores were defined as the top and bottom halves of the distribution of total
scores correspondingly. High scores correlated with a higher level of extraversion, while low
scores correlate with lower levels of extraversion.
Typically, norms groups for objective personality tests like TES would consist of a
sample that is intended to be representative of the demographic composition of the population
EXTRAVERSION 15
being investigated. This sample would ideally be achieved through random sampling and
assignment which is not possible in this case. As a result, tests like the one presented in this
research do not often have norm groups.
The TES is considered to be a qualification level A test. This means that there are no
special qualifications needed to purchase or use these materials.
Initially, a reliability value of approximately .95 was expected. This was because special
care was taken in order to ensure that all of the items on the measure were interconnected. A
reliability value at this level is an indication of excellent reliability and the ability of a test to
contribute to weighty decisions. A test with a reliability value this high could stand on its own
with no need to supplement the test score with additional material (Hogan, 2015). The
coefficient alpha value that was actually obtained for this test was .916. This value, although not
as high as it was expected to be, is still an indication of excellent reliability, and a test that could
stand on its own. In the future, reliability could be increased by incorporating standardized
procedures for the administration of the test to the sample. Instead of allowing the respondents
to take the test home, the test would be administered at a set date and time in a prearranged
room.
A Pearson correlation was computed in order to determine the criterion related validity
for the Test for Extraversion and the Shyness Survey. A negative correlation was found between
extraversion and Antisocial behavior (r=-.044). A negative correlation was also found between
extraversion and shyness (r=-.203). These findings indicated that there was discriminant validity
between extraversion, shyness, and antisocial behavior. This means that there is relatively little
correlation between these three constructs.
EXTRAVERSION 16
Construct validity was examined using the coefficient alpha value that was calculated to
determine the internal consistency of the measure. The coefficient alpha value that was obtained
was r s=.92. This score is an indication of not only a high degree of reliability, but also the
possibility that the measure has a high degree of validity as well. More information would be
needed in order to determine whether or not the test is actually valid however.
Implications
Some of the limitations of this study include the small sample size of only 25 people
which makes it impossible to generalize the results to the entire population of black females on
this construct. Further research will need to be done with a larger sample size in order to garner
better results.
Although this test was found to have both excellent reliability, and a high degree of
construct validity, there is no way to really know whether or not the test is actually measuring
what it intends to measure. These values only indicate that some construct is being measured in
a reliable and valid manner, but does not provide information about whether or not it is
measuring the correct construct. Further information would be needed to actually draw
conclusions on this test, hopefully, this is something that can be explored in the future.
EXTRAVERSION 17
References
Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job
performance: A meta-analysis. Retrieved April 28, 2016, from
http://onlinelibrary.wiley.com/doi/10.1111/j.1744-6570.1991.tb00688.x/abstract
Carlyn, M. (1977). An assessment of the Myers-Briggs type indicator. Journal Of Personality
Assessment, 41(5), 461.
Dentrick, P., & Chibnall, J. T. (2013). Revised NEO personality inventory normative data for
police officer selection (4th ed., Vol. 10). St. Louis, MO: American Psychological
Association. doi:10.1037/a0031800
Harbough, E. R. (2010). The effect of personality styles (Level of Introversion ... Retrieved
April 27, 2016, from
http://www.elon.edu/docs/e
-web/academics/communications/research/vol1no2/06HarbaughEJFall10.pdf
Hogan, T. P. (2015). Psychological testing: A practical introduction (3rd ed.).
Danvers, MA: Wiley & Sons.
Huffman, K. (2011). Psychology in Action (10th ed.). Danvers, MA: John Wiley & Sons.
Kim, H., & Chung, Y. W. (2014). The use of social networking services and their relationship
with the big five personality model and job satisfaction in Korea. Cyberpsychology,
Behavior, And Social Networking, 17(10), 658-663. doi:10.1089/cyber.2014.0109
King, E. B., George, J. M., & Hebl, M. R. (2005). Linking personality to helping behaviors at
EXTRAVERSION 18
work: An interactional perspective. Journal Of Personality, 73(3), 585-608.
doi:10.1111/j.1467-6494.2005.00322.x
The Myers & Briggs Foundation - MBTI® Basics. (n.d.). Retrieved April 26, 2016, from
http://www.myersbriggs.org/my-mbti-personality-type/mbti-basics/
Otonari, J., Nagano, J., Morita, M., Budhathoki, S., Tashiro, N., Toyomura, K., & ... Takayanagi,
R. (2012). Neuroticism and extraversion personality traits, health behaviours, and
subjective well-being: the Fukuoka Study (Japan). Quality Of Life Research, 21(10),
1847-1855. doi:10.1007/s11136-011-0098-y
Salgado, J. F., Moscoso, S., & Alonso, P. (2013). Subdimensional structure of the Hogan
personality inventory. International Journal of Selection & Assessment, 21(3), 277-285.
doi:10.1111/ijsa.12037
Tiwari, T., Singh, A. L., & Singh, I. L. (2009, January). The short-form revised Eysenck
personality questionnaire: A Hindi edition (EPQRS-H). Retrieved April 25, 2016, from
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3016694/
Zhai, Q., Willis, M., O’Shea, B., Zhai, Y., & Yang, Y. (2013). Big Five personality traits, job
satisfaction and subjective wellbeing in China. Retrieved April 27, 2016, from
http://web.a.ebscohost.com.ezproxy.auctr.edu:2051/ehost/pdfviewer/pdfviewer?
sid=b8488d0a-841d-4a66-b79d-a8cbaa43867b@sessionmgr4004&vid=1&hid=4114
EXTRAVERSION 19
Appendix A
Charts/tables
Items initially written Was the item used in the final survey?
I find it difficult to introduce myself to other people
Yes
I do not like being the center of attention Yes
I am quick to get involved in social activities at my new job or school
Yes
I do not find it difficult to introduce myself to other people
Yes
I enjoy meeting new people Yes
I enjoy being the center of attention Yes
I do not find it difficult to open up to new people
Yes
A good book or video is more interesting than a social event
Yes
I consider myself to be reserved Yes
I find it easy to relax when talking in front of other people
Yes
I like to take the lead in social situations Yes
I find small talk gets the conversation going Yes
I like to get to know as many people as possible
Yes
Being around large groups of people makes me comfortable
yes
I feel drained after being around large groups of people
Yes
I do not consider myself to be reserved Yes
I find small talk to be uncomfortable Yes
EXTRAVERSION 20
I feel comfortable around large groups of people
Yes
I find it difficult to open up to new people Yes
I am slow to get involved in social activities at my new job or school
yes
Table 1. Illustrates the number of items that were initially written, as well as the final set of items that were selcted
EXTRAVERSION 21
Appendix B
Data layout
EXTRAVERSION 22
SPSS Syntax/Output
Statistics
Item analysis for the 20 extraversion items
GET FILE='E:\psychometrics\test for extroversion data set.sav'.DATASET NAME DataSet1 WINDOW=FRONT.DATASET ACTIVATE DataSet1.SAVE OUTFILE='E:\psychometrics\test for extroversion data set.sav' /COMPRESSED.CORRELATIONS /VARIABLES=extroversion_1 extroversion_2 extroversion_3 extroversion_4 extroversion_5 extroversion_6 extroversion_7 extroversion_8 extroversion_9 extroversion_10 extroversion_11 extroversion_12 extroversion_13 extroversion_14 extroversion_15extroversion_16 extroversion_17 extroversion_18 extroversion_19 extroversion_20 /PRINT=TWOTAIL NOSIG /STATISTICS DESCRIPTIVES XPROD /MISSING=PAIRWISE.
Descriptive StatisticsMean Std.
DeviationN
extroversion_1 3.30 1.174 20extroversion_2 2.95 1.050 20extroversion_3 3.00 .918 20extroversion_4 3.35 1.182 20extroversion_5 4.00 .858 20extroversion_6 2.85 1.040 20extroversion_7 2.95 .999 20extroversion_8 3.40 .995 20extroversion_9 2.70 1.218 20extroversion_10
3.00 1.170 20
extroversion_11
2.85 1.089 20
extroversion_12
3.65 .933 20
EXTRAVERSION 23
extroversion_13
3.50 .889 20
extroversion_14
3.50 .889 20
extroversion_15
2.95 1.276 20
extroversion_16
2.70 1.081 20
extroversion_17
3.30 1.174 20
extroversion_18
3.15 1.089 20
extroversion_19
2.90 1.071 20
extroversion_20
2.85 1.268 20
Internal consistency reliability: Split half (first half/second half) with Spearman Brown
[DataSet1] E:\psychometrics\test for extroversion data set.sav
EXTRAVERSION 24
[DataSet1] E:\psychometrics\test for extroversion data set.sav
NONPAR CORR /VARIABLES=firsthalftest secondhalftest /PRINT=SPEARMAN TWOTAIL NOSIG /MISSING=PAIRWISE.
EXTRAVERSION 25
Internal consistency reliability: Split half (odd/even) with Spearman Brown
Internal consistency reliability: coefficient alpha
EXTRAVERSION 26
[DataSet1] E:\psychometrics\test for extroversion data set.sav
Item StatisticsMean Std.
DeviationN
extroversion_1 3.30 1.174 20extroversion_2 2.95 1.050 20extroversion_3 3.00 .918 20extroversion_4 3.35 1.182 20extroversion_5 4.00 .858 20extroversion_6 2.85 1.040 20extroversion_7 2.95 .999 20extroversion_8 3.40 .995 20extroversion_9 2.70 1.218 20extroversion_10
3.00 1.170 20
extroversion_11
2.85 1.089 20
extroversion_12
3.65 .933 20
extroversion_13
3.50 .889 20
EXTRAVERSION 27
extroversion_14
3.50 .889 20
extroversion_15
2.95 1.276 20
extroversion_16
2.70 1.081 20
extroversion_17
3.30 1.174 20
extroversion_18
3.15 1.089 20
extroversion_19
2.90 1.071 20
extroversion_20
2.85 1.268 20
EXTRAVERSION 28
EXTRAVERSION 29
Criterion-related validity: convergent/discriminant
[DataSet1] E:\psychometrics\merged data set with peers for validity.sav
Extraversion and shyness
Antisocial behavior and extraversion
[DataSet1] E:\psychometrics\merged data set with peers for validity.sav
EXTRAVERSION 30
EXTRAVERSION 31