See discussions, stats, and author profiles for this publication at: [Link]
net/publication/283264604
Adaptation of Clinical Decision Making in Nursing Scale to Undergraduate
Students of Nursing: The Study of Reliability and Validity
Article in International Journal of Psychology and Educational Studies · January 2015
CITATIONS READS
17 2,633
2 authors:
Aylin Durmaz Aklime Sarıkaya [Dicle]
Dokuz Eylul University Istanbul Aydin University
30 PUBLICATIONS 241 CITATIONS 28 PUBLICATIONS 639 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Investigation of symptoms severity, symptoms clustering and status of interference in the life of patients with primary brain tumors View project
All content following this page was uploaded by Aylin Durmaz on 27 October 2015.
The user has requested enhancement of the downloaded file.
International Journal of Psychology and Educational Studies, 2015, 2 (3), 1-9
International Journal of Psychology and Educational
Studies
[Link]
Adaptation of Clinical Decision Making in Nursing Scale to
Undergraduate Students of Nursing: The Study of Reliability and Validity
Aylin Durmaz Edeer1, Aklime Sarıkaya2
1
Dokuz Eylul University, Turkey, 2İstanbul Sabahattin Zaim Üniversitesi
AR TI CL E IN FO A BS TRA CT
Article History: A clinical decision making skill is essential in the implementation of nursing
Received 06.05.2015
knowledge and reflecting on patient care. The research was planned to measure
Received in revised form
20.08.2015 the reliability and validity of The Clinical Decision Making in Nursing Scale
Accepted 29.08.2015 (CDMNSTr) for undergraduate nursing students from Turkey. This study is a
Available online methodological design. This study was conducted on 210 undergraduate
23.09.2015
students of nursing. For validity; Language – Content Validity and Construct
Validity (Exploratory and Confirmatory Factor Analysis) were examined. For
reliability; CDMNS’s Cronbach’s alpha reliability coefficient, item-total score
correlation coefficients and stability analysis (test-retest) were examined. Item
Content Validity Index and Scale Content Validity Index were calculated as .81
and .83 respectively. Confirmatory factor analyses showed that goodness of fit
indexes were acceptable. Cronbach alpha value of the scale was .78. Item-to-total
score correlation coefficients ranged from .13 to .56. The correlation coefficient
for test-retest was .82. The scale can be used as a valid and reliable measurement
tool to determine the perceptions of Turkish undergraduate students of nursing
regarding to clinical decision making.
© 2015 IJPES. All rights reserved
Keywords: 1
Decision Making, Nursing student, Reliability, Validity
1. Introduction
Decision making skills are fundamental for nurses who must make effective decisions in a complex and
ever-changing healthcare environment (Jenkins, 2001). Nurses are health personnel who analyze the data of
the change in patient’s condition patients’ conditions and determine the priorities; they are also responsible
for clinical decision making in care together with the patient and family (Tanner, 2006). Clinical decision
making defines practicing as the most appropriate, useful and acceptable alternative among the solutions in
order to overcome the problems of the client or patient and his family (Thompson & Dowding, 2002).
Clinical decision making in nursing includes the type of care that comes after the effect of illness on patient
and family. It also includes determining emotional, socio-cultural and economic shortcomings of patient and
family and then using necessary skills to cope with those shortcomings (Tanner, 2006). Briefly, clinical
decision making in nursing means practicing professional nursing knowledge and skills (Jenkins, 1983;
Tanner, 2006).
1Corresponding author’s address: Aylin Durmaz Edeer, Dokuz Eylul University, Faculty of Nursing, Izmir
Telephone: 0232 4124764
Fax: 0232 4124798
e-mail: aylin_durmaz@[Link]
[Link]
© 2014 International Journal of Psychology and Educational Studies (IJPES) is supported by Educational Researches and Publications Association (ERPA)
International Journal of Psychology and Educational Studies 2015,2 (3) 1-9
“Clinical decision making” is one of the basic skills developed during baccalaureate nursing education and
all graduates are expected to be equipped with these skills (American Association of Colleges of Nursing,
2008). The World Health Organization (WHO) has published the golden standards of nursing education and
according to these standards, the development of clinical decision making skills should be provided in
nursing school programs (World Health Organization, 2009). Decision making is required in order to
acquire expertise (Dunphy & Williamson, 2004). It is necessary to determine the perception of nursing
students in clinical decision making, and to develop and evaluate their decision making skills. Therefore,
valid and reliable measurement tools are required to evaluate students’ perceptions in clinical decision
making as well as the way they make decisions. Both at the national and the international level, the number
of measurement tools which evaluate clinical decision making skills is limited. Only the Clinical Decision
Making in Nursing Scale (CDMNS) was found in the study as a measurement tool for evaluating the
perceptions of nursing students in decision making. Adapting the CDMNS to Turkey fulfilled the need at the
national level and provided an opportunity to retest the scale in a different culture at the international level.
The CDMNS is used to identify and to evaluate clinical decision making in nursing. The CDMNS was
developed by Jenkins (1983). The internal reliability of the items used in the scale where the CDMNS was
developed was discussed during a panel with expert educators of undergraduate nursing education and the
items on which a consensus was reached were included. Cronbach’s alpha reliability coefficient of the
original CDMNS internal consistency was found to be .83 and the explanatory factor analysis has showed
that the four-factor structure explain 72.3 % of the total variance (Jenkins, 1983 & 1985).
Byrnes and West (2000) used the scale to evaluate the perceptions of nursing students in clinical decision
making in Australia. The reliability and validity of CDMNS was not examined in their study. Girot (2000)
found that Cronbach’s alpha reliability coefficient was .78 among the Canadian graduate nurses. The validity
in this study was tested by a group of experienced practitioners who were considered to be 'expert decision-
makers' in practice and they established content validity. On the other hand, Baumberger-Henry (2005)
found that Cronbach's alpha coefficient was .81 among the nursing students in the USA and the validity in
this study was not tested. Gorton (2010) used the CDMSN tool to investigate clinical judgment of the nurse
practitioner students and the reliability of the instruments used in this study was evaluated. Cronbach’s
alpha coefficient was .73 for the CDMNS tool and .67 for the CDMNS evaluation and reevaluation subscale.
No study directly evaluating the clinical decision making of undergraduate students in Turkey was found.
However, the evaluative studies on the problem solving processes and critical thinking skills of nursing
students state that these skills would indirectly affect decision making. It is obvious that there is a need for a
valid and reliable measurement tool that would evaluate the clinical decision making skills of nursing
students to help them prepare for professional life. This study was conducted in order to examine the
validity and the reliability of the Turkish version of the CDMNS as a tool for evaluating the perception of
nursing students in clinical decision making.
2. Method
The research method was a scale adaptation study which was structured based on screening model.
2.1. Sample of the Research
The research was conducted in Dokuz Eylul University, school of nursing in 2009. The research sample
comprised 210 undergraduate students of nursing who had previous experience of clinical practice. In
thelight of Tavşancıl’s recommendations; there were 5 to 10 people per item of an instrument (Tavşancıl,
2006) and the sample size of 210 was considered to be sufficient to conduct factor analysis of the CDMNS
which comprised 40 items.
The data were collected with a “Defining Characteristics” form which was composed of three questions and
the “CDMNS” in the classroom environment. 210 undergraduate nursing students participated in the
research. These students completed their clinical practice. The mean age of the students was 21.13 ± 1.07. All
of the students were female. 28.6% of the students (n= 60) were sophomores, 38.1% (n= 80) were in their third
year and 33.3% (n=70) were seniors.
2.2. Instruments
Data were collected by using a Demographic Form and The Clinical Decision Making in Nursing Scale.
2
Aylin Durmaz Edeer
2.2.1. The Clinical Decision Making in Nursing Scale (CDMNS): The original CDMNS was developed by
Jenkins (1983) with nursing students in the USA. This scale describes the perception of the nursing students
in clinical decision making based on self-expression (Jenkins, 2001).
The original CDMNS is composed of 40 items and four subscales. The subscales of the scale are “search for
alternatives or options”, “canvassing of objectives and values”, “evaluation and reevaluation of
consequences”, and “search for information and unbiased assimilation of new information”. Each subscale
is composed of 10 items. 22 items (1, 3, 5, 7, 8, 9, 10, 11, 14, 16, 17, 18, 20, 26, 27,28, 29, 33, 35, 36, 37 and 38) are
written as positive. 18 items (2, 4, 6, 12, 13, 15, 19, 21, 22, 23, 24, 25, 30, 31, 32, 34, 39, and 40) are written as
negative. In this scale, 18 items are inversely scored. Each item of the scale is evaluated through the five-
point likert scale as 5=Always, 4=frequently, 3=occasionally, 2=Seldom, and 1=Never (Jenkins, 1983).
Minimum and maximum points to be taken are 40 and 200 in the whole scale and 10 and 50 in the subscales,
and there is no cutting point. A high score taken from the scale indicates that the perception in decision
making is high, whereas a low score indicates that the perception in decision making is low. The scale is
evaluated through the scores obtained from each subscale and the total scale (Jenkins, 1983; 1985; 2001).
2.3. Data Collection
The researchers were given information about the scale and about how to fill it. The objective of the study
was explained to a total of 216 students. 210 students volunteered to participate in the study and 6 students
refused it. The scales were distributed to the students participating in the study by the researchers. The
students completed the scale.
Each student was asked to write down their self-selected password on the scale both during the first attempt
and the test-retest practice which was conducted 6 weeks later to check the stability of the scale. Thus, it was
possible to gather the data safely by hiding the students’ identities and to match them up. As a result of the
re-test, 109 students (51.9 %) who responded to the scale were taken into consideration, incomplete forms
and forms with mismatching passwords were excluded. Each participant needed approximately 10-15
minutes to complete the scale. Demographic data were self reported by the students and subsequently
obtained from the demographic form.
2.4. Ethical considerations
Ethical approval was obtained from the Ethics Committee of the School of Nursing. During data collection,
the students were informed about the aim of the research and verbal informed consent was obtained from
each participant.
2.5. Data analysis
Data were analyzed by using Statistical Package for Social Sciences (SPSS) version 15.0 and LISREL 8
statistical program software. The Content validity of the Turkish version of CDMNS was tested by
requesting opinions of experts using the Content Validity Index (CVI). The Exploratory Factor Analysis
(EFA) and The Confirmatory Factor Analyses (CFA) were used to determine the construct validity. The
Confirmatory Factor Analyses (CFA) was used to determine the construct validity of the Turkish version of
CDMNS with LISREL 8 statistical program software. In terms of scale reliability, Cronbach’s alpha reliability
coefficient and the item analysis were used to find out the internal consistency of the scale and the subscales.
The stability of the scale was tested by test-retest reliability coefficients.
3. Results
3.1. Descriptive Statistics of CDMNS-Tr
The CDMNS-Tr score mean is 160.82±10.75 and the subscale score means are between 39.78±3.29 and
40.58±3.45. The lowest and the highest scores for CDMNS were 132.00 and 185.0 respectively. The standard
error value of the scale was determined to be .74 whereas the standard error values were between .21 and .25
for the subscales in Table 1.
3
International Journal of Psychology and Educational Studies 2015,2 (3) 1-9
Table 1: Results of the Clinical Decision Making in Nursing Scale and Subscale Analysis (n: 210)
CDMNS and
Mean-SD SE* Median Min Max r+ α≠
Subscale
Search for alternatives
40.58±3.45 .23 41.00 29.00 50.00 .82 .50
or options
Canvassing of
39.78±3.29 .22 40.00 32.00 49.00 .77 .44
objectives and values
Evaluation and
reevaluation of 39.91±3.72 .25 40.00 30.00 49.00 .80 .52
consequences
Search for information
and unbiased
40.54±3.13 .21 41.00 31.00 50.00 .74 .40
assimilation of new
information
The total of CDMNS 160.82±10.75 .74 161.00 132.00 185.00 .78
*SE: Standart errors
+r: Correlation coefficients
≠α: Cronbach’s Alpha Reliability Coefficient
3.2. Validity of the CDMNS-Tr
3.2.1. Linguistic Validity. In order to ensure the language validity of the original CDMNS, language
experts who are familiar with both languages and cultures translated the scale from English to Turkish. The
most suitable expressions were selected from the translated versions of the CDMNS, and a single version of
the scale was created. In order to test whether the Turkish version of the scale provided the same meaning,
the resultant Turkish version of the CDMNS was backtranslated into English, the scale was once more
translated by two different experts who had not seen the English version of the scale and had a good
command of both languages. The items of the back-translated scale were examined and it was seen that the
meanings were close to the original scale. The linguistic validity of the CDMNS was confirmed.
3.2.2. Content Validity. To test content validity, a total of eight experts specializing in nursing education
were asked to give their opinions about the CDMNS whose content validity was confirmed. Each question
in the CDMNS was scored by the experts on a 4-point scale: 1=not relevant, 4=highly relevant. In accordance
with the experts’ recommendations, necessary changes in the items were made. Evaluations of expert
opinions were made with Content Validity Index (Polit & Beck, 2006). The Content Validity Index (CVI) was
calculated both for the items and the scale. The item–CVI was calculated by using the formula of the number
of experts who gave three points (quite relevant) or four points (highly relevant) for each item divided by the
total number of experts. The calculated ratios were then added up and the total ratio was divided by the
total number of items. It was recommended not to have Item Content Validity Index below 0.78 (Polit &
Beck, 2006), if there were six or more experts. For the scale- CVI, each expert was separately evaluated at the
first step. For each expert’s evaluation, the total number of their rating of 3 or 4 (i.e. quite or highly relevant)
was divided by the total number of items. Then the ratios which were calculated for each expert were added
up and divided by the total number of experts. Scale- Content Validity Index was recommended to be 0.80
minimum (Polit & Beck, 2006). The Content Validity Index for Items (I-CVI) and The Content Validity Index
for Scale (S-CVI) were calculated as .81 and .83 respectively.
3.2.3. Pre-application. Linguistic and content validity of the CDMNS was completed. It was piloted on 12
senior students having the characteristics of the study sample. Three students stated that items 14, 20 and 31
were not comprehensible. In accordance with the feedback from these students, necessary changes were
made on the items without altering the meaning. As a result of these revisions in the scale, the final version
of the scale was applied to the whole sampling group.
4
Aylin Durmaz Edeer
3.2.4. Construct Validity. In the adaptation study of the scale, the construct validity was first tested by
exploratory factor analysis and then confirmatory factor analaysis in the same sample group (n:210).
Recently, it has been suggested to perform exploratory and confirmatory factor analysis for similar but
separate samples. It can be considered as a limitation to perform for the same sample in this study. The
results were given separately under different subtitles.
In Exploratory Factor Analysis (EFA), Kaiser-Meyer-Olkin value was the evaluation criteria for sample
adequacy. Kaiser-Meyer-Olkin value was .73. (Bartlett’s Test of Sphericity; X2: 2039,161, df: 780 and p< 0.001).
Seven factor of scale explains 60.8 % of the total variance.
Secondly, confirmatory factor analysis (CFA) with structural equation modeling was conducted to
check the construct validity of the CDMNS-Tr. In CFA, the goodness of fit statistics and modification index
results were examined without any restrictions in the model by adding new connections. The results of the
goodness of fit statistics of the scale were as in the following: [(χ2 (740, N= 210) =1725.02, p=0.000,
RMSEA=.080, S-RMR=.089, GFI=.71, AGFI=.68, CFI =.76] (Table 2). The results of the goodness of fit statistics
of the items that constitute the subscale were as in the following: [(χ2 (734, N = 210) = 1711.93, p=0.000,
RMSEA=.08, S-RMR=.089, GFI=.71, AGFI=.68, CFI=.76].
Table 2: Confirmatory Factor Analysis of Clinical Decision Making in Nursing Scale
Expected CDMNSTr
CFA Model Compatibility Indexes
Values Form
Minimum Fit Function Chi-Square (χ2)
χ2 / df < 5 χ2 / df = 2.3
Degrees of Freedom (df)
Root Mean Squared Error of Approximation (RMSEA) <.08 .08
Standardized Root Mean Square Residual(SRMR) <.08 .089
Comparative Fit Index (CFI) >.90 .76
Goodness of Fit Index (GFI) >.90 .71
Adjusted Goodness of Fit Index (AGFI) >.90 .68
3.3. Reliability of the CDMNS-Tr
3.3.1. CDMNS’s Cronbach’s Alpha Reliability Coefficient. The total Cronbach’s alpha reliability coefficient
of the CDMNS was .78. It was found out to be .50, .44, .52 and .40 in the subscales, respectively (Table 1).
3.3.2. CDMNS’s Item - Total Correlation. Item analysis is the method of assessment of correlation
coefficients between item and total score. CDMNS’s item - total correlation ranged from .13 to .56 for
CDMNS-Tr. All correlation coefficients were statistically significant (P< .05). The items 2, 11, 27 and 28,
respectively were .20 less than the scale item the total correlation. The items which were .20 less than the
scale item total correlation coefficients were 2,11,27 and 28, respectively. These items were about professional
responsibilities and values. There was no increase in the correlation coefficients in the absence of these items.
Cronbach’s alpha coefficient did not change during the analysis in the absence of these items. These items
were excluded from the scale since the total score of the scale; the total score of the sub-scale and sub-scale
total correlation were high and acceptable. The total scale score and the total subscale score correlation were
between .74 and .82 (Table 1).
3.3.3. CDMNS’s Stability Analysis. CDMNS’s stability was examined by comparing the test-retest
Pearson correlation coefficients. There was not a significant difference in the scores for the CDMNS-Tr
between test–retest total scores and the subscale total scores (P> 0.05). Correspondingly, the CDMNS’s test-
retest total score correlation coefficient was .82 and subscale total score correlation coefficients were .66, .56,
.63, .67, respectively and this difference was significant (p=0.000).
5
International Journal of Psychology and Educational Studies 2015,2 (3) 1-9
4. Discussion
In this study, we tested the reliability and validity of the CDMNS for the Turkish culture in a sample of
nursing students. This study examined the linguistic validity, content validity, construct validity and
reliability of CDMNS in nursing students within Turkish culture.
4.1. Validity
Linguistic validity. Translation of a scale from its original version to the target language and its back
translation was the most commonly used methods (Aksayan & Gözüm, 2002). Translators’ knowledge and
experience have a great influence on the result. Therefore, translators who know cultures and who have a
good command of both languages should be selected (Aksayan & Gözüm, 2002). The CDMNS’s language
understandability was evaluated. For this reason, the scale was translated into Turkish by two people
knowing both languages and cultures well and its back translation was made by two other people who
know both languages and cultures well but had not seen the scale before. While preparing the Turkish form
of the scale, particular attention was paid to ensure that the statements were suitable for the Turkish
language structure and had the same cultural connotations (Hilton & Skrutkowski, 2002). The linguistic
validity was confirmed.
Content validity. The aim of content validity is that experts decide whether items of a scale represent the
construct planned to be measured and create a scale including meaningful items (Eser, 2007; Ercan & Kan,
2004). It is recommended that expert opinion regarding the content validity should be requested from three
specialists minimum and ten specialists maximum (Polit & Beck, 2006). In this study, to test the content
validity of the scale, a total of eight experts specialized in nursing education were asked to comment on
whether the items of the CDMNS were appropriate.
CVI was used to determine whether the experts agreed (Polit & Beck, 2006). CVI is computed two ways; item
and scale CVI. Item -CVI was computed for each item and should be greater than .78. Scale -CVI was
computed for the all the items of scale and should be greater than .80 (Polit & Beck, 2006). In this study I-CVI
and S-CVI values of the CDMNS-Tr was found acceptable. (Polit&Beck, 2006). The values indicated a
consensus among experts concerning items of the CDMNS-Tr.
Construct validity. KMO value was .73 in explanatory factor analysis. KMO values between .70 and .79 were
considered to be good values. This showed the sample size to be sufficient to carry out the factor analysis
(Akgül and Çevik, 2005).
Confirmative factor analysis (CFA) was usually used to develop scales, revise the scales or evaluates
construct validity (Jackson, Gillaspy, Purc-Stephenson, 2009). Confirmative factor analysis (CFA) is used to
give information about the construct validity. In order to observe the construct validity of the scale
adaptation, the similarity to the original scale factor construct was checked by CFA (Dimitrov 2010; Şimşek,
2007). CFA showed that the goodness of fit statistics were [(χ2 (df= 740, N=210) = 1725.02, p=0.000, X2/df: 2.3
RMSEA=.080, S-RMR=.089, GFI=.71, AGFI=.68, CFI=.76]. The analysis showed that the Chi square value (χ2)
was significant. A high χ2 value was common in the majority of the samples. Therefore, the calculation was
done by dividing χ2 value by degrees of freedom (df). This ratio being five or lower indicates that the model
has acceptable goodness of fit (Şimşek, 2007). As the value obtained by dividing the CDMNS’s χ2 into df was
2.3, the model had acceptable goodness of fit. In addition to χ2 values, CFA examined many other goodness
of fit statistics. The most common ones among these are GFI, AGFI, CFI, RMSEA and SRMR. GFI, AGFI and
CFI values above .90 (Schreiber, Nora, Stage, Barlow, King, 2006; Ullman 2006; Şimşek, 2007; Jackson,
Gillaspy, Purc-Stephenson, 2009), and RMSEA and SRMR values below .80 are indicators of acceptable
goodness of fit (Schreiber, Nora, Stage, Barlow, King, 2006; Şimşek, 2007). However, in this study GFI, AGFI
and CFI were all below .90 and hence, the goodness of fit was not as expected. On the other hand, RMSEA
and SRMR were both .80 and were within the acceptable limits, indicating that the factor construct is similar
to that of the original scale. The model constructed according to the subscales also has the same
characteristics (Table 2). In the modifications suggested for this model, items were associated with the
subscales as; item 3 with the subscale of ‘search for information and unbiased assimilation of new
information’, items 11, 15 and 29 with the subscale of ‘search for alternatives or options’, item 36 with the
subscale of ‘canvassing of objectives and values’ and item 40 with the subscale of ‘Evaluation and
6
Aylin Durmaz Edeer
reevaluation of consequences’. The suggested modification analyses were not conducted as these suggested
items were closely correlated and all the items had a theoretical relationship in general.
CFA showed that the items 2, 11, 25, 27, 28, 30 and 31 were statistically insignificant (p>0.05, t value <1.96).
The correlation coefficients of these items was less than 0.20. When these items were excluded, CFA
goodness of fit values did not manifest a significant change [(χ2 (df= 528, N=210)=1088.74, p=0.000,
RMSEA=.084, S-RMR=.090, GFI=.73, AGFI=.70, CFI=.76]. Therefore, the items were not excluded from the
model. These items were related with professional values, patients’ and families’ values which were the
important components of clinical decision making and it was needed to reevaluate the relationship between
these items and before mentioned values. Correlation coefficients of confirmatory factor analysis were under
acceptable limits in this study and this meant that the scale needed to be reevaluated.
4.2. Reliability
The standard error of the scale is presented in Table 1 as a measure to support its reliability. The low
standard error of the scale means that its reliability is high, whereas a high standard error indicates low
reliability (Tavşancıl, 2006). The low standard errors of both the CDMNS total and the subscales strengthen
the reliability of the measurement tool.
Cronbach’s alpha reliability coefficient is an important indicator of reliability. The total Cronbach’s alpha
reliability coefficient of the CDMNS was .78. It was determined to be .50, .44, .52 and .40, respectively in the
subscales. Cronbach’s alpha reliability coefficient ranging between .60 and .80 indicates that the scale is
notably reliable, whereas a value between .40 and .60 specifies that the scale has low reliability (Özdamar,
2004; Tavşancıl, 2006). According to these criteria, the CDMNS is reliable; nevertheless the subscales has low
reliability. The original CDMNS Cronbach’s alpha reliability coefficient was 0.83. Jenkins (1985) did not
report reliability for any of the subscales of the original CDMNS. Whereas the study of Baumberger-Henry
(2005) found it out to be 0.81 and the subscales were .53, .57, .58 and .51. Cronbach’s alpha reliability
coefficients of the scale and the subscales were similar. The results indicate that the scale items are consistent
and they constitute a whole. The scale should be applied as a whole. The sub-scales of CDMNS have low
Cronbach's alpha reliability coefficients. Therefore, the sub-scales of CDMNS are not used alone. Even
though the reliability coefficients of total scale were under acceptable limits in this study, they were at lower
yet acceptable levels in subscales and this was important as it showed that the measurement tool needed to
be further improved.
One of the methods that show the internal consistency of the scale is item analysis. It is observed that the
CDMNS’s item-total correlation coefficients vary between 0.13 and 0.56 and these values are statistically
significant (p<0.05). The correlation coefficients of the scale-total score and the subscale-total score were .74
and .82, respectively (Table 1). Although it was not presented in the findings, the subscale item-total
correlation coefficients were within the range of .22-.57. Jenkins (1983) did not provide the item correlations
of the original scale. Except for the 4 items that are below the acceptable levels in the item analysis, it is seen
that 40 items are consistent among them and thus constitute a whole. Given that the correlation coefficients
below .20 indicate a weak relationship these items were suggested to be excluded from the scale, but this is
not a strict rule. When the item correlations are below .20 and the items are deleted, it is recommended to
check the change in Cronbach’s alpha coefficient. Only if Cronbach’s alpha coefficient increases when an
item is excluded, can the item be deleted. It should be noted whether these items could be distinctive
(LoBiondo-Wood & Haber, 2005; Özdamar, 2004; Şencan, 2005). It was seen that the items 2, 11, 27 and 28
scored below 0.20 when the CDMNS’s item-total correlations were observed. When the items related to the
perception in professional responsibility and values are excluded, it is seen that Cronbach’s alpha coefficient
does not change. Scale-total score, subscale-total score and subscale-item-total score correlation coefficients
are high and above the acceptable levels and this indicates that these items should not be excluded.
Therefore, it was concluded that these items support the scale and do not change the reliability.
Test-retest values in the adapted scale indicate the consistency of the measurement tool from practice to
practice and its stability through time. In the light of the recommendation to assess stability over a 2 to
6weeks period (Tabachnick&Fidell; 1996), we administered the retest approximately 6 weeks after the initial
administration.
7
International Journal of Psychology and Educational Studies 2015,2 (3) 1-9
It is suggested that both measurement results should be similar (Gözüm & Aksayan, 2003; LoBiondo-Wood
& Haber, 2005). The similarity of the CDMNS’s test-retest total scores and subscale scores (p>0.05), and their
intermediate and high correlation (p= 0.000) show that the scale is consistent and stable.
Limitations
Although it is an important tool to evaluate the clinical decision making, the cronbach’s alpha reliability
coefficient of the subscales and the results of the goodness of fit statistics of the scale are low. This is a
limitation.
5. Conclusion
In conclusion, it is observed that the Turkish version of the Clinical Decision Making in Nursing Scale
(CDMNSTr) is a reliable and valid tool for examining the perceptions of the Turkish undergraduate students
of nursing in clinical decision making. However, the low correlation coefficients in some items observed
during the item analysis are related to different interpretations of language, culture and professional values.
Although the original meaning was obtained in the language validity of the scale, it is possible that the
adapted society has diverse cultural conceptual schemes. Therefore, it is suggested that the concepts put
forward by the items be qualitatively examined and that the items be re-arranged. In general, the results
indicate that the CDMNS would be useful in determining the perceptions of undergraduate nursing
students in decision making after their first clinical practice at the national level. The data gathered through
this scale could provide the basis for developing the perceptions of the students in decision making and
improving the nursing curriculum in order to help students gain this skill.
References
Akgül, A., Çevik, O. (2005). Faktör analizi, İstatistiksel analiz teknikleri SPSS'de işletme yönetimi
uygulamaları. [Link]ı. Ankara, Kişisel Yayınlar.
Aksayan, S., Gözüm, S. (2002). Kültürlerarası ölçek uyarlaması için rehber I: Ölçek uyarlama aşamaları ve dil
uyarlaması. Hemşirelik Araştırma Geliştirme Dergisi, 4(1):9-14.
[Link]
American Association of Colleges of Nursing (AACN). (2008). The essentials of baccalaureate education for
professional nursing practice. Retrieved from [Link]
(19 Kasım 2013)
Baumberger-Henry, M. (2005). Cooperative learning and case study: does the combination improve
students’ perception of problem-solving and decision making skills?. Nurse Education Today, 25(3), 238–
246. doi:10.1016/[Link].2005.01.010
Byrnes, M., West, S. (2000). Registered nurses' clinical reasoning abilities: a study of self perception. The
Australian Journal of Advanced Nursing, 17(3), 18-23.
Dimitrov, D.M. (2010). Testing for factorial invariance in the context of construct validation. Measurement
and Evaluation in Counseling and Development, 43(2),121-149. doi:10.1177/0748175610373459
Dunphy, B. C., Wıllıamson S. L. (2004). In pursuit of expertise toward an educational model for expertise
development. Advances in Health Sciences Education, 9(2), 107–127.
Ercan, İ., Kan, İ. (2004). Ölçeklerde güvenirlik ve geçerlik. Uludağ Üniversitesi Tıp Fakültesi Dergisi), 30(3):211-
216. [Link]
Eser, E. (2007). Sağlıkla ilgili yaşam kalitesi ölçeklerinin kültürel uyarlaması. 2. Sağlıkta Yaşam Kalitesi
Kongresi, Kongre Öncesi Kurslar Kitabı, İzmir.
Girot, E.A. (2000). Graduate nurses: critical thinkers or better decision makers?. Journal of Advanced Nursing,
31(2), 288-297. doi: 10.1046/j.1365-2648.2000.01298.x
8
Aylin Durmaz Edeer
Gorton, K.L.(2010). An Investigation into the relationship between critical thinking skills and clinical judgment in the
nurse practitioner student. Published Doctor of Philosophy Dissertation, College of Natural and Health
Sciences, School of Nursing, Program of Nursing Education, University of Northern Colorado.
Gözüm, S., Aksayan, S. (2003). Kültürlerarası Ölçek Uyarlaması İçin Rehber II: Psikometrik Özellikler ve
Kültürlerarası Karşılaştırma. Hemşirelikte Araştırrma Geliştirme Dergisi, 5(1), 3-14.
[Link]
Hilton, A., Skrutkowski, M. (2002). Translating instruments into other languages: development and testing
processes. Cancer Nursing, 25(1), 1-7.
Jackson, D. L., Gillaspy, J. A. Jr., & Purc-Stephenson, R. (2009). Reporting practices in confirmatory factor
analysis: An overview and some recommendations. Psychological methods, 14(1), 6-23.
Jenkins, H. (1983). Perceptions of decision making among baccalaureate nursing students as measured by the
clinical decision making in nursing scale. doctors of thesess, University of Maryland.
Jenkins, H. (1985). A research tool for measuring perceptions of clinical decision making. Journal of
Professional Nursing, 1(4), 221-229.
Jenkins, H.M. (2001). Cinical decision making in nursing scale. In: Waltz, C.F., Jenkins, L.S. (Eds.),
Measurement of Nursing Outcomes volume:1 measuring nursing performance in practice, education
and research, Springer Publishing Company, 33-37.
LoBiondo-Wood, G., Haber, J. (2005). Nursing research in Canada. Methods, critical appraisal and
utilization. Elseiver Mosby, Canada, 327–347.
Ozdamar, K. (2004). Güvenirlik ve soru analizi, Paket programlar istatistiksel veri analizi. 4. Baskı, Eskişehir,
Kaan Kitabevi, 661-676.
Polit, D.F., Beck, C.T. (2006). The content validity index: are you sure you know what’s being reported?
critique and recommendations, Research in Nursing & Health, 29(5), 489–497. doi: 10.1002/nur.20147
Schreiber, J.B., Nora, A., Stage F.K., Barlow E.A., King J. (2006). Reporting structural equation modeling and
confirmatory factor analysis results: a review. Journal of Educational Research, 99(6), 323-337.
doi:10.3200/JOER.99.6.323-338
Sencan, H. (2005). Sosyal ve davranışsal ölçümlerde güvenirlik ve geçerlik. Ankara, 250-252.
Simşek, Ö.F. (2007). Yapısal eşitlik modellemesine giriş temel ilkeler ve lisrel uygulamaları. Ankara,
Ekinoks,11-49.
Ullman, J.B. (2006). Structural equation modeling: Reviewing the basics and moving forward, Journal of
Personality Assessment, 87(1), 35-50. doi:10.1207/s15327752jpa8701_03.
Tanner, C.A. (2006). Thinking like a nurse: a research-based model of clinical judgment in nursing. Journal of
Nursing Education, 45(6), 204-211.
Tabachnick, B.G. (1996). Fidell, L.S. Using Multivariate Statistics. Northridge, CA: HarperCollins.
Tavşancıl, E. (2006). Ölçme ile ilgili temel kavramlar. Tutumların ölçülmesi ve SPSS ile veri analizi. 3. Basım,
Ankara, Nobel yayın dağıtım, 17-19.
Thompson, C., Dowding, D. (2002). Clinical decision making and judgement in nursing. London, Churchill
Livingstone.
World Health Organization (WHO). (2009). Nursing & Midwifery Human Resources for Health, Global
standards for the initial education of professional nurses and midwives. Retrieved June 22, 2010, from
[Link]
View publication stats