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

net/publication/283666181

Development of the self-directed learning skills scale

Article in International Journal of Lifelong Education · October 2015

DOI: 10.1080/02601370.2015.1091393

CITATIONS READS

54 11,678

2 authors:

Yildizay Ayyildiz Leman Tarhan

Dokuz Eylul University Dokuz Eylul University
14 PUBLICATIONS 295 CITATIONS 101 PUBLICATIONS 2,530 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Yildizay Ayyildiz on 25 December 2017.

The user has requested enhancement of the downloaded file.

 International Journal of Lifelong Education

ISSN: 0260-1370 (Print) 1464-519X (Online) Journal homepage: http://www.tandfonline.com/loi/tled20

Development of the self-directed learning skills

scale

Yildizay Ayyildiz & Leman Tarhan

To cite this article: Yildizay Ayyildiz & Leman Tarhan (2015) Development of the self-
directed learning skills scale, International Journal of Lifelong Education, 34:6, 663-679, DOI:
10.1080/02601370.2015.1091393

To link to this article: https://doi.org/10.1080/02601370.2015.1091393

Published online: 15 Oct 2015.

Submit your article to this journal

Article views: 693

View related articles

View Crossmark data

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tled20

Download by: [Dokuz Eylul University ] Date: 25 December 2017, At: 03:31
 INT. J. OF LIFELONG EDUCATION, 2015
VOL. 34, NO. 6, 663–679, http://dx.doi.org/10.1080/02601370.2015.1091393

Development of the self-directed learning

skills scale

YILDIZAY AYYILDIZ and LEMAN TARHAN

Dokuz Eylul University, Turkey
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

The purpose of this study was to develop a valid and reliable scale for assessing high
school students’ self-directed learning skills. Based on a literature review and data
obtained from similar instruments, all skills related to self-directed learning were identi-
fied. Next, an item pool was prepared and administered to 255 students from various
high schools. To test the suitability of the gathered data, exploratory factor analysis was
performed. The results revealed that there were correlations between the items, factor
analysis could be conducted and nine factors were obtained. A confirmatory factor analy-
sis (CFA) was performed concerning the quality of the factor structure. The results of the
CFA confirmed the nine-factor solution. The final version of the scale has a nine-factor
structure and includes a total of 40 items. This instrument uses a five-point Likert-type
scale and was termed the Self-Directed Learning Skills Scale (SDLSS).

Keywords: self-directed learning; self-directed learning skills; scale development

Introduction

Literature review on self-directed learning

In the teaching process, principal decisions are made regarding such issues as
how learners learn, how much they have learned, what their deficiencies are,
how they focus on the lesson, when and from whom they can obtain help and
how they understand the aims of the learning processes. Traditionally, most of
these decisions are made by the teacher. During active learning, which has
recently emerged, the responsibility of the learning process belongs to the lear-
ner and students make the decisions for themselves. Thus, learners attempt to
learn by their own endeavour rather than implementing decisions made by their
teachers, parents or more informed individuals (Brandes & Ginnis, 1986).
The concept of self-directed learning was proposed by Dewey and Lindeman
in the 1900–1930s. Lindeman (1926) asserted that highest value in adult

Yildizay Ayyildiz received her PhD in Chemistry Education. She is a lecturer in Torbali
Vocational School at Dokuz Eylul University. Her primary research interests include
science and chemistry education, student-centred teaching and learning, constructivist
approach. Correspondence: Science Faculty, Department of Chemistry, Dokuz Eylul
University, 35160, Buca, Izmir, Turkey. Email: [email protected].

Leman Tarhan is a professor in the Faculty of Science at Dokuz Eylul University. Her pro-
fessional interests include chemistry, biochemistry, science and chemistry education, tea-
cher education. Correspondence: Science Faculty, Department of Chemistry, Dokuz Eylul
University, 35160, Buca, Izmir, Turkey. Email: [email protected]

© 2015 Taylor & Francis

 664 YILDIZAY AYYILDIZ AND LEMAN TARHAN

education is the learner’s experience rather than required classes. Later, the
concept was elaborated on and implemented by Knowles, Tough and Houle
(Maeroff, 2003). Knowles (1975) argued that self-directed learning involves the
abilities to decide with or without help from other people, to determine what
one needs for learning, to express learning achievements clearly and implicitly,
to select and implement appropriate learning strategies and to assess learning
outcomes (Oladoke, 2006). Knowles (1975) explained the learning steps imple-
mented by an individual with self-directed learning skills as follows: determining
the learning needs, expressing the learning aims clearly and implicitly, deter-
mining the learning materials, selecting and implementing appropriate learning
strategy and assessing learning outcomes.
The concept of self-directed learning is related to the concepts of self-regula-
tion, self-sufficiency and self-control. Students with self-directed learning skills
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

require control, regulation, internal and external motivation and success during
learning activities and experiences (O’Shea, 2003). Students who control their
own learning experiences can easily transfer what they have learnt in today’s
scientific context in which interdisciplinary studies are of great importance and
thus perform greater number of operations during their own learning processes
(Boyer & Kelly, 2005).
The development and strengthening life-long learning attitudes in students
might vary according to other attitudes, skills and behaviours that the students
seek to acquire. The development of a life-long learning attitude involves the
acquisition of new skills, such as information gathering by the students and the
development of some specific attitudes and notions about learning (Parkinson,
1999). Another issue that is related to life-long learning is self-directed learning.
Self-directed learners have specific characteristics, attitudes and skills that
include the attitude that learning is a personal responsibility, an approach that
involves engaging challenging problems and eagerness to learn. Life-long learn-
ing characteristics include being motivated, independent, self-disciplined and
self-confident. Life-long learning skills are comprised of basic studying skills and
time management (Bolhuis, 1996).
When the literature on the classification of self-directed learning skills has
been examined, titles that involve attitudes of personal responsibility for learn-
ing, challenging problems, willingness for learning and basic studying and time
management skills have drawn attention (Boynak, 2004). Raemdonck (2006)
mentioned the subjects of determining targets, monitoring and assessment, and
selecting and pursuing strategies. Additionally, Long (2007) grouped self-
directed learning skills into the two main areas of personal characteristics and
general skills that include cognitive properties, such as determining targets and
processing information. It has been emphasised that a self-directed learner
should possess skills that include determining learning aims, defining appropri-
ate learning resources, selecting appropriate learning strategies, distinguishing
what is important and what is not, forming a whole from different resources,
time management and monitoring the achievement of learning outcomes and
the efficiency of study habits (Towle & Cottrell, 1996). In contrast, Knapper and
Cropley (2000) asserted that the following are important skills for self-directed
learners: assessing their own learning; being more active than passive students;
learning in both academic and non-academic contexts; learning from peers,
teachers and monitors; gathering information from different lessons and
 SELF-DIRECTED LEARNING SKILLS SCALE 665

disciplines when needed; and implementing different learning strategies in

different situations.
Adams (2006), Cornford (2002), Crowther (2004) and Koç (2007) argued
that self-directed learners should possess characteristics, such as a desire for con-
tinuous learning, a feeling of responsibility for their own learning, the ability of
learning how to learn, reading for understanding, basic number skills, oral and
written communication skills, knowledge of and skills related to information
technologies, a wide inventory of strategies to ensure efficient learning, self-de-
velopment skills, high-level thinking skills, such as problem solving and critical
thinking, self-regulated learning skills, research skills, social skills (i.e. not having
difficulties in initiating and maintaining interpersonal relations, being capable
of teamwork and cooperative learning, etc.).
The concept of self-directed learning has been mentioned in many studies in
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

various disciplines and instruments have been developed to evaluate this skill
and administered to students. The most prominent of such instruments was cre-
ated by Guglielmino, Guglielmino, and Long (1987) and these researchers
aimed to determine the relation between readiness for self-directed learning
and performance in a professional context. In this study, the researchers
administered the Self-Directed Learning Readiness Scale (SDLRS) to a sample
group and tested the significance of variables including administration level,
gender, age, educational status, religious sect, job performance, creativity level
and problem solving skills required for the job. The results revealed that the
scores of the individuals who worked in jobs that required higher levels of cre-
ativity and problem solving skills and those with greater levels of education were
significantly higher than those of the other participants.

Measurements related to self-directed learning

When the most significant scales that measure the self-directed learning skills of
students are investigated, the first to be encountered is the SDLRS, which was
developed in 1977 by Guglielmino. This scale was developed to comprehend the
influences of the individual characteristics, skills and motivations of self-directed
learners and contains 41 items that are rated on a five-point Likert scale. The
SDLRS (Guglielmino, 1977) includes sub-factors of openness to learning possi-
bilities, self-concept as an efficient learner, independence and enterprise in
learning, awareness of self-responsibility for learning, the love of learning,
creativity, positive prospects for the future, basic education skills and problem
solving skills.
Another scale termed the Self Direct Learning Scale toward Science and
Technology Course was developed by Aydede and Kesercioğlu (2009) and com-
prises 25 items. This scale has two sub-factors. The first is ‘planning self-directed
learning in science and technology courses’. The factor load values of this sub-
factor, which comprises 15 items, vary between .45 and .58 and its Cronbach’s
alpha coefficient is .91, which corresponds to 19.8% of the total variance. The
second sub-factor is ‘confidence in self-directed learning in science and technol-
ogy courses’. The factor load values of this sub-factor, which comprises 10 items,
vary between .46 and .72 and its Cronbach’s alpha confidence coefficient is .78,
which corresponds to 15.7% of total variance of the scale. The Cronbach’s alpha
coefficient for the entire scale is .86.
 666 YILDIZAY AYYILDIZ AND LEMAN TARHAN

In the study conducted by Fisher, King, and Tague (2001), the development
of the SDLRS for Nursing Education was realised in two steps. First, the Delphi
technique was used via a panel of 11 nurse training experts to evaluate the valid-
ity of the structure and the content of the items expressing the self-directed
learning readiness. The opinions of all panel members on each item of the
SDLRS for Nursing Education were taken via a Likert scale for independent
evaluation. In the second step, the scale was administered to 201 nursing stu-
dents and the homogeneity of the scale was ensured after validity and reliability
studies. Eventually, the SDLRS for Nursing Education was developed with 41
items that comprised three sub-factors of self-management, learning eagerness
and self-control features.
The Self-Rating Scale of Self-Directed Learning was developed by Williamson
(2007) for use in higher education to determine which skills would be required
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

for life-long learners. In the process of the development of the scale, the studies
of researchers who studied the subject of self-directed learning, such as Gugliel-
mino, Knowles, Candy, Hiemstra and Brookfield, were investigated initially and
a draft scale comprising 75 items was formed. Next, it was deduced via the facil-
itation of the views of experts via the Delphi technique that 15 items were not
suitable and some of them might remain in the scale after corrections. In its lat-
est version, the scale is comprised of 60 items and includes the following five
sub-factors: awareness (self-evaluation), learning strategies, learning activities,
assessment and interpersonal skills.
Guglielmino’s SDLRS is a popular instrument used to assess individual
characteristics, skills and motivations of self-directed learners. She identified
eight factors which define attitudes, values and abilities of adult learners associ-
ated with readiness for self-directed learning in her instrument. These factors
form the basis of the SDLRS. Even though Guglielmino’s SDLRS has been devel-
oped and also translated into 22 languages, it can only be used for a certain fee.
The Self Direct Learning Scale toward Science and Technology Course devel-
oped by Aydede and Kesercioğlu was structured around only two factors, plan-
ning self-directed learning and confidence in self-directed learning in science
and technology courses. That is to say that this scale does not have a conceptual
framework for understanding self-directed learning. It was developed for only
Turkish language, not translated into other languages. Additionally, the
researchers developed the scale for only science and technology courses, hence
the name.
The SDLRS for Nursing Education developed by Fisher, King and Tague
allows only nurse educators to diagnose students’ attitudes, abilities and per-
sonality characteristics necessary for self-directed learning. It means that this
scale is limited to perform on undergraduate nursing students. Otherwise, the
scale was structured around only three factors, self-management, learning eager-
ness and self-control features. It shows that this scale does not have an adequate
content for all self-directed learning skills, like The Self Direct Learning Scale
toward Science and Technology Course.
Williamson’s Self-Rating Scale of Self-Directed Learning was developed to
assess self-directed learning behaviour that is different from the simple measur-
ing of perceptions and readiness for self-directed learning. Using this scale is
also limited with the undergraduate nursing student population, like the scale
developed by Fisher, King and Tague.
 SELF-DIRECTED LEARNING SKILLS SCALE 667

Furthermore, although, there have been several instruments of students’ self-

directed learning skills in the literature, these instruments suffer from lacking
all the skills related to self-directed learning and there are no scales for high
school students. Therefore, a new alternative instrument that measures high
school students’ all self-directed learning skills described in the literature with
improved validity and reliability is needed. Accordingly, the purpose of this
study was to develop a valid and reliable scale that assesses high school students’
self-directed learning skills. This five-point Likert-type scale was termed the Self-
Directed Learning Skills Scale (SDLSS).

Method
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Participants

The 47-item pilot version of the scale was administered to 255 students from dif-
ferent high schools in Izmir, Turkey. During the selection process, a list of high
schools in Izmir was first obtained. Next, seven high schools were randomly
selected for the validation study. The data were collected from one regular high
school (N = 42:21,21), three Anatolian high schools (N = 92:47,45), one science
high school (N = 41:20,21) and two private high schools (N = 80:40,40). There
were a total of 128 tenth and 127 eleventh grade students.

Measurement development

In this study, the scales for Self-Directed Learning that have been previously pre-
pared and are still being administered both in Turkey and abroad were exam-
ined (Aydede & Kesercioğlu, 2009; Fisher et al., 2001; Guglielmino, 1977;
Williamson, 2007). In this respect, the following aspects were studied:

• how many sub-factors comprised each scale,

• which self-directed learning skills the items comprising the scales aimed
to measure,
• the associations of the items of all the scales with each other and the
determination of the number of repetitions of each skill,
• how the skill groups that the items addressing the same learning out-
comes could be grouped.

As a result of the analyses conducted based on these criteria, the pros and
cons of each scale were presented. Later, the sub-factors of the scales were deter-
mined and the items in all of the scales were grouped in terms of the skills they
addressed based on the inter-item correlations. Finally, the frequencies of each
skill were also calculated.
In the present study, we gained insight from the results of the frequency
analyses of each determined skill about how many items in the SDLSS to be
developed would reflect the skills. Considering all of these parameters, the
SDLSS was developed via the formation of 49 items, among which 10 items are
negative, and was completed with 47 items based on the exclusion of 2 items
and the correction of some of the items based on the experts’ opinions.
 668 YILDIZAY AYYILDIZ AND LEMAN TARHAN

One of the logical means of testing the scope validity of a scale is expert opin-
ion. Experts are expected to assess the items in the draft form of the scale in
terms of scope validity (Büyüköztürk, 2007). For this purpose, experts’ opinions
on the items in the SDLSS were elicited from five lecturers and four high school
teachers. Moreover, the opinions of three high school students were elicited
regarding the comprehensibility of the scale items. Increases in the scope
validity of the scale were attempted based on making corrections based on the
opinions and suggestions received from the experts. The experts assessed the
initial scale in terms of the following aspects:

• comprehensibility of the items,

• which skill each item measured,
• the appropriateness of the items to define self-directed learning skills,
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

• opinions about and criticisms toward the items.

The scale was finalised according to the experts’ opinions for this age group,
and the students were asked to indicate the option most appropriate for them-
selves on the following five-point Likert type scale: ‘Strongly agree’, ‘Agree’, ‘Par-
tially Agree’, ‘Disagree’ and ‘Strongly Disagree’. The results were graded as
scores of one to five for the negative to positive opinions.

Results

In the context of this study, the kappa statistics were calculated to measure the
reliability of the comparative fit between the researcher and the experts in terms
of the inter-rater reliability of the scale. With these statistics, the fit between the
items designed by the researcher and the items defined by the experts regarding
self-directed learning skills was examined. Landis and Koch (1977) defined the
different intervals of kappa values with regard to the level of fit. Values equal to
or greater than .75 indicate perfect fits, values equal to or below .40 indicate
weak fits, and values between .40 and .75 indicate mid-level fit (Fleiss, Levin, &
Paik, 2003). When the averages of the kappa coefficients between each expert
and the researcher for SDLSS were considered, the Cohen’s kappa coefficient
value was found to be .71. In other words, the fit was found to be a mid-level fit
and close to perfect fit between the researcher and the experts regarding the
SDLSS.

Item analysis

The basic assumption of Likert type scales is that each item measures a single
attitude (Tavşancıl, 2005). Therefore, to present the evidence for the reliability
of the scale, an item analysis was conducted. This analysis revealed that the
Cronbach’s alpha for the internal consistency coefficient was .84 and items 8,
11, 16, 18, 24, 28 and 38 were removed from the scale, which caused an increase
in this value. The remaining 40 items were again enumerated and all the analy-
sis results are presented in terms of this new enumeration. Testing the differ-
ences between the item average points of the lower and upper 27% of the
groups which are formed according to the total points of the data collection
 SELF-DIRECTED LEARNING SKILLS SCALE 669

tool using unrelated t-tests, is an additional method of item analysis. To

determine the discrimination of the items, the total points of the students in
the initial administration of the scale were calculated and placed into two
groups of the upper and lower 27%. Later, the discrimination of the items was
identified using t-tests for independent groups with the SPSS software package.
The obtained t values revealed the discrimination power of the items in the scale
and the discrimination power increased with increasing t value (Tavşancıl,
2005). These analyses revealed that there were statistically significant differences
(p < .001) between the upper and lower group averages for all items. According
to Büyüköztürk (2007), the statistically significant difference observed between
the groups as a result of the analysis can also be expressed as an indication of
the internal consistency of the data collection tool.
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Exploratory factor analysis

Factor analysis is a method in which factor reduction is performed and the

dependence structures are abolished by determining common factors for the
scale (Tavşancıl, 2005). Factor analysis is used to determine whether the items
in a scale can be divided in a smaller number of factors that exclude each other.
Thus, an attempt is made to identify factors that comprise items measuring the
same factor (Balcı, 2005).
Forty items remained following the item analysis and an exploratory factor
analysis (EFA) was conducted to ensure the structural validity of the scale. Struc-
tural validity is realised by providing explanations of the measured property over
the specific structures or concepts (Cohen, Manion, & Morrison, 2002). EFA is
one of the most widely used statistical methods in the social sciences (Fabrigar,
Wegener, MacCallum, & Strahan, 1999). When conducting an EFA, it is impor-
tant to make decisions regarding the estimation method, the number of factors
to retain, the rotation method and the method for calculating scores (DiStefano,
Zhu, & Mı̂ndrilã, 2009; Parmet, Schechtman, & Sherman, 2010). Principal
components analysis, principal axis factoring, maximum likelihood factor analy-
sis and minimum residual factor analysis are four of the most popular estimation
methods in EFA. Principal components analysis, which was used in this study,
refers to the principal components model in which items are assumed to be the
exact linear combinations of factors. Principal axis factoring uses squared multi-
ple correlations as initial estimates of the communalities. These communalities
are entered into the diagonals of the correlation matrix before factors are
extracted from this matrix. Maximum likelihood produces parameter estimates
that are the most likely to have produced the observed correlations, if the sam-
ple is from a multivariate normal population. Minimum residual factor analysis
extracts factors from the correlation matrix ignoring the diagonal elements.
Moreover, the varimax rotation technique was used for factor rotation in this
study. Varimax rotation technique is one of the better factoring methods and
usually provides a clear separation of factors. The technique extracts factors
using normalised factor loadings during the iterations (Yong & Pearce, 2013).
At end of the analysis, the Kaiser-Meyer-Olkin (KMO) coefficient was found to
be .81 and the Bartlett test was found to be statistically significant
(χ2 = 5331.374, df = 780, p = .000 < .001) (table 1).
 670 YILDIZAY AYYILDIZ AND LEMAN TARHAN

Table 1. KMO and Bartlett test results for the SDLSS

Kaiser-Meyer-Olkin’s measure of sampling adequacy .806
Bartlett’s test of sphericity Chi-squared test (χ2) 5331.374
df 780
Sig. (p) .000

The KMO results greater than .60 and the statistical significance of the Bar-
tlett test indicated that the data were appropriate for factor analysis (Büyüköz-
türk, 2007).
When the results of the EFA were examined, the total variance values
revealed that the 40 items in the scale could be grouped into 9 factors with
eigenvalues greater than 1. The total variance of the scale explained by these 9
factors was 65%. The common variances of the 40 items were observed to be
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

between .42 and .93. Moreover, after the properties of the items under the fac-
tors were examined in terms of the data in the literature, a scree plot was con-
sidered and it was decided that the scale could have nine factors (figure 1).
When the rotated factor loadings were examined, all items were found to be
greater than .50 and there were no items that produced higher values in two or
more factors. The maximum values for each factor are presented in table 2.
As a consequence of the factor analysis, the SDLSS was comprised of 9 factors
and a total of 40 items with 8 negative items. Based on the data in the literature
regarding the skills to be measured by each factor, the following factor names
were given:

• attitude towards learning (ATL),

• learning responsibility (LR),
• motivation and self-confidence (MS),
• ability to plan learning (PL),
• ability to use learning opportunities (ULO),

Figure 1. Scree plot of the exploratory factor analysis of the SDLSS

 SELF-DIRECTED LEARNING SKILLS SCALE 671

Table 2. The largest rotated factor loadings for the SDLSS

Components
1 2 3 4 5 6 7 8 9
Item 1 .943
Item 10 .874
Item 40 .864
Item 15 .850
Item 34 .842
Item 13 .902
Item 28 .893
Item 35 .879
Item 17 .695
Item 21 .643
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Item 16 .566
Item 6 .873
Item 33 .785
Item 36 .782
Item 27 .739
Item 7 .736
Item 38 .901
Item 25 .886
Item 4 .873
Item 8 .780
Item 29 .735
Item 11 .716
Item 22 .646
Item 12 .609
Item 32 .721
Item 5 .689
Item 20 .665
Item 24 .654
Item 9 .598
Item 30 .744
Item 2 .736
Item 14 .708
Item 26 .705
Item 31 .779
Item 19 .738
Item 39 .724
Item 37 .635
Item 18 .868
Item 23 .834
Item 3 .741

• ability to manage information (MI),

• ability to apply learning strategies (ALS),
• assessment of learning process (ALP),
• evaluation of learning success/results (ELS).

Confirmatory factor analysis

To test the accuracy of the nine factors identified by the EFA, confirmatory
factor analysis (CFA), which has become quite popular recently, was used (Ang
& Huan, 2006; Barrett, 2007; Kraimer, Seibert, & Liden, 1999; Lievens & Anseel,
 672 YILDIZAY AYYILDIZ AND LEMAN TARHAN

2004). The Lisrel 8.51 software (Cheng & Chan, 2003) was used for this CFA.
The covariance matrices were used for the tested model, and the fit statistics
and modification results were examined in CFA (Büyüköztürk, Akgün, Özkah-
veci, & Demirel, 2004; Çolakkadıoğlu & Güçray, 2007).
In the CFA process, the chi-square fit test, goodness-of-fit index (GFI),
adjusted goodness-of-fit index (AGFI), comparative fit index (CFI), normalised
fit index (NFI), non-normalised fit index (NNFI) and root mean square error of
approximation (RMSEA) values were calculated (Akın & Çetin, 2007; Sanders
et al., 2005; Tosun & Irak, 2008).
Generally, χ2/df ratios equal to or below 3, CFI, NNFI, NFI, GFI, and AGFI
values greater than .90 and RMSEA significance levels below .06 show that the
factor structure is well fit (Heubeck & Neill, 2000; Hoe, 2008; Hu & Bentler,
1999; Kahn, 2006; Kelloway, 1998; Kline, 1998; Sanders et al., 2005). In general,
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

GFI, AGFI and CFI values between .80 and .90 show that the structure is suitable
for a good fit and values greater than .90 indicate adequately good fit (Corral &
Calvete, 2000).
RMSEA values below .05 indicate perfect fits, values between .05 and .08 indi-
cate acceptable fit, and values between .08 and .10 indicate average fit (Hoe,
2008). The CFA revealed that χ2 = 747.92, df = 704; RMSEA = .016, χ2/df = 1.06,
NFI = .86, NNFI = .99, CFI = .99, GFI = .87 and AGFI = .85. Consequently, it
could be argued that the fit indices obtained by testing the accuracy of the
model for the SDLSS suggested a perfect fit. The path diagram for the analysis
is given in figure 2.
After the item analysis, EFA and CFA of the scale, Cronbach’s alpha values
for the sub-factors of the scale were calculated to determine the reliability of the
scale and the items classified under each factor are listed in table 3 with the cor-
responding item numbers. The Cronbach’s alpha reliability for the entire scale
was .86; therefore, the SDLSS was found to have a reliable structure as presented
in the Appendix 1.

Discussion

This study aimed to develop a valid and reliable scale that assesses high school
students’ self-directed learning skills. As a result of analysis, the Cohen’s kappa
coefficient value measuring the reliability of the comparative fit between the
researcher and the experts in terms of the reliability of the scale was found to
be .71. On the other hand, the Cronbach’s alpha reliability coefficient for the
SDLSS was found to be .86. However, these findings of the study are limited to
255 students from different high schools in Izmir, Turkey and may not ordinar-
ily be generalised to other students. Therefore, all statistical results obtained
from this study need to be validated in larger samples.
The findings of this study indicated that the SDLSS provided valid and reli-
able scores for the examination of high school students’ self-directed learning
skills. This scale was also found to be a practical tool for teachers to use because
the administration of this scale does not require long periods of time and it is
not difficult to interpret the results.
In spite of strengths of the existing scales in the literature, there are also
weaknesses in the conceptual foundation which comprise a basis for the scales.
 SELF-DIRECTED LEARNING SKILLS SCALE 673
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Figure 2. Confirmatory factor analysis model for the SDLSS

Table 3. Cronbach’s alpha values for reliability

Factor Cronbach’s
number Factor name Item number alpha
1 Attitude towards learning 19, 31, 37, 39 .725
2 Learning responsibility 3, 18, 23 .776
3 Motivation and self-confidence 2, 14, 26, 30 .750
4 Ability to plan learning 5, 9, 20, 24, 32 .746
5 Ability to use learning 4, 25, 38 .949
opportunities
6 Ability to manage information 6, 7, 27, 33, 36 .846
7 Ability to apply learning 1, 10, 15, 34, 40 .936
strategies
8 Assessment of learning process 13, 16, 17, 21, 28, 35 .870
9 Evaluation of learning success/ 8, 11, 12, 22, 29 .764
results
 674 YILDIZAY AYYILDIZ AND LEMAN TARHAN

Since the existing scales measure a homogeneous rather than a multi-factorial

construct, we aimed to develop a new scale measuring a multi-factorial construct.
CFA supported the claim that self-directed learning is a multi-factorial construct
consisting of different processes. The final version of the SDLSS includes 40
items in nine factors. These factors were ATL, LR, MS, ability to PL, ability to
ULO, ability to MI, ability to ALS, ALP and ELS/results. The reliability coeffi-
cients of the factors were also found to be reasonably high and within the range
of .725 for ATL and .949 for ability to ULO.
Zimmerman, Bonner, and Kovach (1996) proposed that every student can
become a successful learner via the use of self-regulatory strategies. Therefore,
teachers should teach students how to be self-regulated learners. However,
teaching students to be strategic learners involves more than the acquisition of
new strategies. It requires the implementation of strategies in new tasks and the
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

adjustment of those strategies regarding changes in environmental, behavioural

and personal factors. Teachers should make students responsible for their learn-
ing and guide them in this learning process. Additionally, Pape, Bell, and Yetkin
(2003) argued that strategy instruction should be individualised because differ-
ent students have difficulties in different tasks. Some students might experience
difficulty analysing the tasks, while others might struggle regarding the effective
monitoring of their performance. Therefore, teachers could benefit from this
instrument to determine each student’s strengths and weaknesses.
Comparing the SDLSS to other scales, the SDLSS developed in this study will
allow teachers to match their instructional design with student’s readiness for
self-directed learning. Grow (1991) outlines a staged self-directed learning
model where learners advance through stages of increasing self-direction.
Through their method of teaching and level of control, teachers can help or
hinder the learner’s development through the stages. Consequently, the SDLSS
will assist high school teachers in the diagnosis of student learning needs in
order for the teacher to implement teaching strategies that will best suit the stu-
dents. This scale will also provide valuable data for curriculum development.
Furthermore, since the SDLSS was not developed in order to measure
self-directed learning skills of students in a specific field of education, it can
potentially be used in all education fields. Furthermore, the development of a
new scale will provide to overcome the problems associated with the use of the
scales to be addressed. Moreover, the SDLSS developed in this study can be used
without purchase. Considering all the results from the validity and reliability
studies, it can be asserted that the SDLSS can be used to measure the self-di-
rected learning skills of high school students.
Evidence of scope validity has been established by the development of the
SDLSS items from the literature, assessment by experts’ opinions and testing
with EFA. However, additional research is required to provide further evidence
of scope validity. There are two limitations to the present study. First, the sample
was not population-based, which might limit the generalisability to other popula-
tions. Second, although the data were collected from seven different high
schools, the sample was not large. These findings need to be validated in larger
and cross-cultural samples. A criticism of the SDLSS is poor reliability and an
inability to replicate its factor structure across different racial groups. Research
is required to confirm the factor structure of the current scale when applied to
different racial groups.
 SELF-DIRECTED LEARNING SKILLS SCALE 675

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding
This work was supported by The Scientific and Technological Research Council of Turkey [Project
Number TUBITAK-109K574].

References

Adams, J. (2006). Community matters in China. In E. Hannum & B. Fuller (Eds.), Research in sociology
of education (Vol. 15, pp. 15–42). Social Organization of Childhood in Developing Countries. Amster-
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

dam: Elsevier.
Akın, A., & Çetin, B. (2007). Başarı yönelimleri ölçeği: Geçerlik ve güvenirlik çalışması [Achievement
goal orientations scale: The study of validity and reliability]. Eg˘itim Araştırmaları Dergisi, 26, 1–
12.
Ang, R. P., & Huan, V. S. (2006). Academic expectations stress inventory: Development, factor analy-
sis, reliability, and validity. Educational and Psychological Measurement, 66, 522–539.
Aydede, M. N., & Kesercioğlu, T. (2009). Fen ve teknoloji dersine yönelik kendi kendine öğrenme
becerileri ölçeğinin geliştirilmesi [Development of Self Direct Learning Scale toward Science
and Technology Course]. Çukurova Üniversitesi Eg˘itim Fakültesi Dergisi, 36, 53–61.
Balcı, A. (2005). Sosyal bilimlerde araştırma [Research in social sciences]. Ankara: Pegem Akademi
Yayıncılık.
Barrett, P. (2007). Structural equation modelling: Adjudging model fit. Personality and Individual Dif-
ferences, 42, 815–824.
Bolhuis, S. (1996, April). Towards active and self-directed learning. Preparing for lifelong learning, with refer-
ence to Dutch secondary education. Paper presented at the Annual Meeting of the American
Educational Research Association, New York.
Boyer, N., & Kelly, M. (2005). Breaking the institutional mold: Blended instruction, self-direction,
and multi-level adult education. International Journal of Self-Directed Learning, 2(1), 1–17.
Boynak, F. (2004). Bilgisayar destekli devre tasarımı dersi uygulaması [Application of computer aided
circuit design course]. The Turkish Online Journal of Educational Technology, 3, 61–66.
Brandes, D., & Ginnis, P. (1986). A guide to student-centred learning. Oxford: Blackwell.
Büyüköztürk, Ş. (2007). Sosyal bilimler için veri analizi el kitabı, istatistik, araştırma deseni, SPSS uygula-
malari ve yorum [Manual data analysis for the social sciences, statistics, research design and
interpretation SPSS applications]. Ankara: Pegem Akademi Yayıncılık.
Büyüköztürk, Ş., Akgün, Ö. E., Özkahveci, Ö., & Demirel, F. (2004). Güdülenme ve öğrenme strateji-
leri ölçeğinin Türkçe formunun geçerlik ve güvenirlik çalışması [The validity and reliability
study of the Turkish version of the motivated strategies for learning questionnaire]. Kuram ve
Uygulamada Eg˘itim Bilimleri, 4, 207–239.
Cheng, S. T., & Chan, A. C. M. (2003). The development of a brief measure of school attitude. Ed-
ucational and Psychological Measurement, 63, 1060–1070.
Cohen, L., Manion, L., & Morrison, K. (2002). Research methods in education. London: Routledge/Fal-
mer.
Çolakkadıoğlu, O., & Güçray, S. S. (2007). Ergenlerde karar verme ölçeği’ni Türkçe’ye uyarlama
çalışması [Turkish adaptation study of adolescent decision making questionnare]. Eg˘itim Ara-
ştırmaları Dergisi, 26, 61–71.
Cornford, I. R. (2002). Learning-to-learn strategies as a basis for effective lifelong learning. Interna-
tional Journal of Lifelong Education, 21, 357–368.
Corral, S., & Calvete, E. (2000). Machiavellianism: Dimensionality of the mach IV and its relation to
self-monitoring in a Spanish sample. The Spanish Journal of Psychology, 3, 3–13.
Crowther, D. (2004). Limited liability or limited responsibility. In D. Crowther & L. Rayman-Bacchus
(Eds.), Perspectives on Corporate Social Responsibility (pp. 42–58). Aldershot: Ashgate.
DiStefano, C., Zhu, M., & Mı̂ndrilã, D. (2009). Understanding and using factor scores: Considera-
tions for the applied researcher. Practical Assessment, Research & Evaluation, 14(20), 1–11.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of
exploratory factor analysis in psychological research. Psychological Methods, 4, 272–299.
Fisher, M., King, J., & Tague, G. (2001). Development of a self-directed learning readiness scale for
nursing education. Nurse Education Today, 21, 516–525.
 676 YILDIZAY AYYILDIZ AND LEMAN TARHAN

Fleiss, J. L., Levin, B., & Paik, M. C. (2003). Statistical methods for rates and proportions (3rd ed.). Hobo-
ken, NY: Wiley.
Grow, G. (1991). Teaching learners to be self-directed: A stage approach. Adult Education Quarterly,
41, 125–149.
Guglielmino, L. M. (1977). Development of the self-directed learning readiness scale (Doctoral dissertation).
University of Georgia, Athens, GA.
Guglielmino, P. J., Guglielmino, L. M., & Long, H. B. (1987). Self-directed learning readiness and
performance in the workplace. Higher Education, 16, 303–317.
Heubeck, B. G., & Neill, J. T. (2000). Confirmatory factor analysis and reliabılıty of the mental health
inventory for Australian adolescents. Psychological Reports, 87, 431–440.
Hoe, S. L. (2008). Issues and procedures in adopting structural equation modelling technique. Jour-
nal of Applied Quantitative Methods, 3, 76–83.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Con-
ventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Jour-
nal, 6(1), 1–55.
Kahn, J. H. (2006). Factor analysis in counseling psychology research, training, and practice: Princi-
ples, advances, and applications. The Counseling Psychologist, 34, 684–718.
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Kelloway, E. K. (1998). Using lisrel for structural equation modeling: A researcher’s guide. Thousand Oaks,
CL: Sage.
Kline, R. B. (1998). Principal and practice of structural equation modeling. New York, NY: The Guilford
Press.
Knapper, C., & Cropley, A. J. (2000). Lifelong learning in higher education (3rd ed.). London: Kogan
Page.
Knowles, M. S. (1975). Self-directed learning: A guide for learners and teachers. Englewood Cliffs, NJ: Pren-
tice Hall/Cambridge.
Koç, G. (2007). Yaşam boyu öğrenme [Lifelong learning]. In Ö. Demirel (Ed.), Eg˘itimde yeni yo¨nelimler
[New trends in education] (pp. 209–222). Ankara: Pegem Akademi Yayıncılık.
Kraimer, M. L., Seibert, S. E., & Liden, R. C. (1999). Psychological empowerment as a multidimen-
sional construct: A test of construct validity. Educational and Psychological Measurement, 59, 127–
142.
Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data.
Biometrics, 33, 159–174.
Lievens, F., & Anseel, F. (2004). Confirmatory factor analysis and invariance of an organizational citi-
zenship behaviour measure across samples in a Dutch-speaking context. Journal of Occupational
and Organizational Psychology, 77, 299–306.
Lindeman, E. C. (1926). The meaning of adult education. New York, NY: New Republic.
Long, H. B. (2007). Themes and theses in self-directed learning. International Journal of Self-Directed
Learning, 4(2), 1–18.
Maeroff, G. I. (2003). A classroom of one: How online learning is changing our schools and colleges. New
York, NY: Palgrave Macmillan.
Oladoke, O. A. (2006). Measurement of self directed learning in online learners. (Doctoral dissertation).
Capella University, Minnesota, MN.
O’Shea, E. (2003). Self-directed learning in nurse education: A review of the literature. Journal of
Advanced Nursing, 43, 62–70.
Pape, S. J., Bell, C. V., & Yetkin, İ. E. (2003). Developing mathematical thinking and self-regulated
learning: A teaching experiment in a seventh-grade mathematics classroom. Educational Studies
in Mathematics, 53, 179–202.
Parkinson, A. (1999, November). Developing the attribute of lifelong learning. Paper presented at the
Annual Meeting of 29th ASEE/IEEE Frontiers in Education Conference, Puerto Rico.
Parmet, Y., Schechtman, E., & Sherman, M. (2010). Factor analysis revisited—How many factors are
there? Communications in Statistics—Simulation and Computation, 39, 1893–1908.
Raemdonck, I. (2006). Self-directedness in learning and career processes: A study in lower-qualified employees
in Flanders (Doctoral dissertation). Ghent University, Ghent.
Sanders, R. D., Allen, D. N., Forman, S. D., Tarpey, T., Keshavan, M. S., & Goldstein, G. (2005). Con-
firmatory factor analysis of the neurological evaluation scale in unmedicated schizophrenia.
Psychiatry Research, 133, 65–71.
Tavşancıl, E. (2005). Tutumların o¨lçülmesi ve SPSS ile veri analizi [Attitude measurement and data analy-
sis with SPSS]. Ankara: Nobel Yayın Dağıtım.
Tosun, A., & Irak, M. (2008). Üstbiliş ölçeği-30’un Türkçe uyarlaması, geçerliği, güvenirliği, kaygı ve
obsesif-kompülsif belirtilerle ilişkisi [Adaptation, validity, and reliability of the metacognition
questionnaire-30 for the Turkish population, and its relationship to anxiety and obsessive-com-
pulsive symptoms]. Türk Psikiyatri Dergisi, 19, 67–80.
Towle, A., & Cottrell, D. (1996). Self directed learning. Archives of Disease in Childhood, 74, 357–359.
Williamson, S. N. (2007). Development of a self-rating scale of self-directed learning. Nurse Researcher,
14, 66–83.
 SELF-DIRECTED LEARNING SKILLS SCALE 677

Yong, A. G., & Pearce, S. (2013). A beginner’s guide to factor analysis: Focusing on exploratory factor
analysis. Tutorials in Quantitative Methods for Psychology, 9, 79–94.
Zimmerman, B. J., Bonner, S., & Kovach, R. (1996). Developing self-regulated learners: Beyond achievement
to self-efficacy. Washington, DC: American Psychological Association.

Appendix 1. Self-Directed Learning Skills Scale

Dear students,
This questionnaire has been developed to measure your learning skills. The answers you provide
are of great importance to the assessment of your self-directed learning skills.
After reading each sentence carefully, please mark the appropriate option from among the
options of Strongly Agree, Agree, Partially Agree, Disagree and Strongly Disagree with an (X) to indi-
cate how much you agree or disagree with the statement in the sentence.
In this study, your surname will be symbolically defined and your personal information will be
kept confidential.
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

Thank you for your contribution.

Name—Surname: Sex: Female() Male()

School: Grade:

Please mark how much you

agree or disagree with the Strongly Partially Strongly
statement in the sentence Agree Agree Agree Disagree Disagree
1. I take notes about important
points when learning a new
subject
2. I believe that I can learn a
lesson, no matter how it is
complicated
3. I should use the internet for
learning purposes, instead of
having a good time
4. I make use of different learning
strategies depending on the
properties of the subject I am
going to learn
5. I can solve the problems I
encounter during learning based
on cause and effect relationship
6. I have difficulty relating the
information I have learned in
the lessons to the daily life
7. My friends say that I suggest
interesting new ideas while
discussing the learning process

(Continued)
 678 YILDIZAY AYYILDIZ AND LEMAN TARHAN

(Continued).
Please mark how much you
agree or disagree with the Strongly Partially Strongly
statement in the sentence Agree Agree Agree Disagree Disagree
8. The result of an examination is
not an indicator of my learning
achievement
9. I organise my study hours by
making plans
10. I underline the important parts
while reading a text
11. I am aware that the knowledge
that I obtain when I study
immediately before the
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

examination is not permanent

12. If I can relate the new concepts
to old knowledge, the learning is
successful
13. I question the information in the
books I make use of
14. If I am motivated for learning,
any distracting factors do not
sidetrack me from my objective
15. I pay attention to establish
relations between concepts when
I learn a subject
16. After each learning process, I
think about what I should do to
be more successful
17. During each learning process, I
question myself regarding
whether I have made use of the
internet for my purposes
18. I hold myself responsible for my
learning
19. I would like my hobbies during
my leisure time to be didactic
20. I must know clearly and
implicitly the objectives of the
new subject to be learnt
21. After each lesson I question
whether I used the course
materials adequately and
systematically
22. I always assess my achievements
in the exercises/homework I
completed
23. To learn a new subject without
difficulty, I should learn related
previous subjects well
24. Generally, I try to finish my
homework at the last moment
25. When I want to learn a new
subject, I know which learning
resource I should use

(Continued)
 SELF-DIRECTED LEARNING SKILLS SCALE 679

(Continued).
Please mark how much you
agree or disagree with the Strongly Partially Strongly
statement in the sentence Agree Agree Agree Disagree Disagree
26. I begin to worry that I could not
solve the problems that I
encounter
27. I cannot establish accurate
hypotheses about the event or
problems in the subjects that I
have learnt
28. I believe that active participation
in the learning process ensures
the permanency of my
Downloaded by [Dokuz Eylul University ] at 03:31 25 December 2017

knowledge
29. After each learning process, I
assess whether I achieved the
objective and outcomes I
identified at the beginning
30. Instead of feeling despair when I
encounter a difficult subject, I
think about what I should do
31. While planning a new day, I
prioritise time for learning
32. I review the previous knowledge
that forms the basis for the new
subject when I start to learn
something new
33. I can produce alternative
methods to reach solutions when
I solve a problem
34. I have difficulty using different
learning strategies in the
learning process
35. After each learning process, I
assess which of the learning
resources I used was more
efficient
36. Generally, I have difficulty in
integrating information I
obtained from different
resources
37. I believe in the importance of
playing an active role in learning
38. I have difficulty accessing the
information I seek in an
equipped library
39. The important thing is not what
I learn, but whether I’ve got a
passing grade
40. I motivate myself by thinking
about the outcome I will obtain
at the end of a learning process

View publication stats