Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

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

net/publication/324776458

Students’ Moral Behaviour Inventory Development and Validation: A Rasch


Analysis

Chapter · January 2018


DOI: 10.1007/978-981-10-8138-5_15

CITATIONS READS
0 517

3 authors:

Sainah Limbasan Mei-Teng Ling

1 PUBLICATION   0 CITATIONS   
Universiti Malaysia Sabah (UMS)
18 PUBLICATIONS   3 CITATIONS   
SEE PROFILE
SEE PROFILE

Vincent A Pang
Universiti Malaysia Sabah (UMS)
70 PUBLICATIONS   139 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

STEM needs assessment View project

article View project

All content following this page was uploaded by Sainah Limbasan on 06 September 2019.

The user has requested enhancement of the downloaded file.


1
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Students' Moral Behaviour Inventory Development And Validation: A Rasch Analysis

Sainah Limbasan1, Mei-Teng Ling2, Vincent Pang3

Faculty of Psychology and Education, Universiti Malaysia Sabah


1,2,3

1
saina.sl74@yahoo.com, 2lingmeiteng@gmail.com, 3pvincent@ums.edu.my

Corresponding author:

Sainah Limbasan

Faculty of Psychology and Education

Universiti Malaysia Sabah

Jalan UMS

88450 Kota Kinabalu

Sabah, Malaysia

Phone: +06 01118555195

saina.sl74@yahoo.com
2
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Abstract

In this study, an instrument was developed to measure the level of students' moral

behaviour based on the Malaysian Moral Education Syllabus. Samples are from 60

secondary school students; aged from 14 to 17 years old, from a middle school in the

Tuaran district, Sabah. The Students' Moral Behavior Inventory had 108 items, with a four-

point rating scale (never, seldom, often, always) questionnaire that assessed seven learning

areas (self-development, family, environment, patriotism, human rights, democracy and,

peace and harmony) This paper reports the analysis of the quality of each of the items in the

instrument using Rasch Model. Person reliability (0,97) and item reliability (0.95) are

excellent. The infit and outfit mean square for all the items were between 0.50 and 1.50,

except for 10 items that needed to be checked and modified. The principal component

analysis residual showed that there is no noticeable secondary dimension. Summary

statistics show good reliability and separation while the person separation and reliability are

weak. The threshold values observed from the structure calibration were in increasing

order. The lower person reliability might be due to poor sample targeting and a narrow

range of the respondents’ endorsements. Therefore, a bigger sample size with a wider range

and more items are needed for future study. The stability of Students' Moral Behavior

Inventory needs to be tested again after the modification of the misfit items.

Keywords: Rasch analysis; rating scale, measurement, Moral Education, moral

behaviour
3
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Background of Study

Morality is defined as cognition, behaviour and emotions related to the ability to judge right and

wrong. It determine whether to accept or reject a behaviour, and guides emotions that motivate

intentions and actions consistently (Kochanska, 1997; and Kohlberg, 1984). Moral is a value of

choice of an individual, which can also be owned and shared within a culture, religion and

society (Vishalache, 2012). Morality should not be measured by solely grading cognitive

development. (Hawley and Geldhof, 2012) but should involve emotional development and

transformation of the students through their ability to develop their sense of morality. The

curriculum of moral education for primary school consists of 12 core values while the secondary

school includes seven areas of learning (self-development, family, environment, patriotism,

human rights, democracy and, peace and harmony) which consist of 36 values (Ministry of

Education, 2000). Moral education curriculum emphasizes spirituality, family, environmental,

social relationship and humanity for a holistic development of students. In addition, the focus on

moral education is an attempt to establish moral and spiritual strength through the experiences

and values gain from religion, tradition and culture. There are seven areas of learning in the

moral education syllabus of secondary schools that form the basis for nurturing good values

among students. These learning areas focus on efforts to foster students’ spiritual and moral

strengths through the appreciation and the practice of actually utilizing moral values and moral

principles.

The moral education syllabus in Malaysia emphasizes the shaping of the individuals’

good character and high moral standards. Moral education is a subject in the form of a

programme that educates students to become a respectful and responsible citizens. This is

achieved through inculcating, appreciating, reasoning and practicing certain identified values.
4
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Poor morality and personality of the youngsters today have caused considerable worries to the

society, implying that not all the values inculcated in school are accepted and practiced by

students in real life (Ezhar et al., 2007; Zawin, 2010). The level of practice and appreciation of

moral values among the students is still at a low to medium level (Mumtazah and Nurizan, 2009;

Nik Zaharah, 2007). This might lead them to make decisions that are immoral when faced with

conflicts in daily life. According to the Malaysian Moral Education Syllabus on moral behavior

can be seen through seven values (Figure 1).

Figure 1: Values in Malaysian Moral Education Syllabus to Measure Moral Behavior

Research Questions And Objectives

In general, the assessment of a student's morality is through test papers. However, students who

show outstanding performance in the moral education examination do not necessarily practice

good values in life. Similarly, students who are less successful in moral education tests may not

necessarily do things that contradict moral values. Moral education should not be just about

memorizing the values but to actually apply the moral behavior in a real world. This study aimed
5
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

to check the suitability of each of the items in each sub-construct, and to check the scale

function of Students' Moral Behaviour Inventory developed by the researchers.

Methodology

The analyses were conducted using Rasch Rating Scale Model (Andrich, 1978). The 60 students

were taken as sample from a middle school in the Tuaran district, Sabah, Malaysia. The ages of

the students ranged from 14 to 17 years old. The instrument used in this study was a self-

developed questionnaire based on the Malaysian Moral Education Syllabus. The Students' Moral

Behaviour Inventory (SMBI) consisted of 108 items, with a four-point rating scale (never,

seldom, often, always) questionnaire that assessed seven learning areas (self-development,

family, environment, patriotism, human rights, democracy and, peace and harmony) (Table 1).

Table 1: Learning Area and Number of Items

Number of
Learning Area Sub-learning Area Total items
items
Believing in God 3
Trustworthiness 3
Self-esteem 3
Responsible 3
Virtuous 3
Values of self- Tolerance 3
36
development Independence 3
Hardworking 3
Love 3
Justice 3
Rationale 3
Moderation 3
Love of family 3
Respect and obedient to family members 3
Values of family 12
Maintaining tradition of family 3
Responsibility to family 3
6
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Love and appreciate the environment 3


Harmony between human and
Values of 3
environment 12
environmental
Environmental Sustainability 3
Sensitive to environmental issues 3
Love the country 3
Values of
Loyalty to King and country 3 9
patriotism
Willing to sacrifice for the country 3
Protect the rights of children 3
Respect women's rights 3
Values of human Protecting workers' rights 3
15
rights Respect for the rights of the
3
underprivileged
Protecting consumer rights 3
Compliance with regulations and laws 3
Freedom of expression 3
Values of
Freedom of religion 3 15
democracy
Engaging in development of the country 3
The attitude of openness 3
Living together peacefully 3
Values of peace
Mutual assistance and cooperation 3 9
and harmony
Mutual respect between countries 3
Total items 108

Results

Summary statistics

Table 2

Item Reliability and Separation Statistic

Raw Model Infit Outfit


Count Measure
score error MNSQ ZSTD MNSQ ZSTD

Mean 369.2 120.0 .00 .14 .99 -.2 1.00 -.1

S.D. 36.4 .0 .70 .01 .29 2.2 .31 2.0

MAX. 428.0 120.0 2.55 .17 2.28 7.7 2.19 5.8

MIN. 228.0 120.0 -1.30 .13 .60 -3.8 .62 -3.0


7
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Real RMSE .15 ADJ. SD. .68 Separation 4.54 Person reliability .95
Model
.14 ADJ. SD. .68 Separation 4.79 Person reliability .96
RMSE
S.E of Item Mean .07

The item reliability is 0.95 while the item separation is 4.54 (Table 2). An item separation of

more than three and item reliability of more than 0.90 imply that the person sample is enough to

confirm the items’ difficulty hierarchy (Linacre and Wright, 2012).

Table 3

Person Reliability and Separation Statistic

Raw Model Infit Outfit


Count Measure
score error MNSQ ZSTD MNSQ ZSTD

Mean 332.3 108.0 1.28 .16 1.02 -.2 1.00 -.3

S.D. 38.6 .0 1.00 .06 .36 2.9 .38 2.9

MAX. 430.0 108.0 6.15 .72 2.25 7.4 2.28 7.6

MIN. 238.0 108.0 -.62 .14 .40 -6.2 .28 -6.2

Real RMSE .18 ADJ. SD. .99 Separation 5.48 Person reliability .97
Model
.17 ADJ. SD. .99 Separation 5.87 Person reliability .97
RMSE
S.E of Person Mean = .09

The person reliability is 0.97 while the person separation is 5.48. Person separation of more than

two and person reliability of more than 0.80 imply that the instrument is sensitive enough to

distinguish between high and low performers (Linacre and Wright, 2012).
8
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Principal Component Analysis of Residuals

Table 4: Principal Component Analysis of Residuals

Table of Standardized Residual variance (in Eigenvalue units)


Empirical Modeled
Total raw variance in observations = 165.6 100.0% 100.0%
Raw variance explained by measures = 57.6 34.8% 34.3%
Raw variance explained by persons = 20.5% 12.4% 12.2%
Raw Variance explained by items = 37.1 22.4% 22.1%
Raw Variance explained (total) = 108.0 65.2% 100.0% 65.7%
Unexplained variance in 1 contrast
st
= 6.5 3.9% 6.0%
Unexplained variance in 2 contrast
nd
= 5.4 3.3% 5.0%
Unexplained variance in 3rd contrast = 4.2 2.5% 3.9%
Unexplained variance in 4 contrast
th
= 3.8 2.3% 3.5%
Unexplained variance in 5 contrast
th
= 3.6 2.2% 3.3%

Table 5 : Contrast Plot

Loading Measure Entry Item Loading Measure Entry Item


0.59 -0.93 A 73 HA04 -0.56 2.55 a 53 AS05
0.56 -1.3 B 81 HA12 -0.47 1.24 b 96 DM12
0.55 -1.3 C 91 DM07 -0.45 1.02 c 68 PT08
0.54 -0.51 D 74 HA05 -0.43 1.02 d 11 PD11
0.5 -0.93 E 79 HA10 -0.33 0.53 e 35 PD35
0.45 -0.55 F 94 DM10 -0.33 0.56 f 87 DM03
0.43 0.01 G 75 HA06 -0.32 0.53 g 69 PT09
0.39 -0.73 H 92 DM08 -0.32 1.48 h 55 AS07
0.39 -0.17 I 71 HA02 -0.29 1.63 i 76 HA07
0.39 -1.03 J 62 PT02 -0.28 0.42 j 4 PD04

Based on the PCAR analysis in Table 4, the raw variance explained by measures are less than

40% and the Eigenvalue of 1st contrast (6.5) has the strength of around seven items, which is

more than 3. The above may indicate a probable secondary dimension (Linacre, 2003). However,
9
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

looking at the Contrast Table (table 5), there is no items above 0.6 contrast loading, nor items

below -0,6 contrast loading. Therefore, there is no noticeable sub-groups that may contribute to

the secondary dimension.

Wright Map

Figure 2: Wright Map


10
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

The Wright map shows that most of the items are easier to be endorsed, per sub-group. The mean

of item measure is lower than mean of person measure. This indicates, on average, respondents

agree to most items. There are many items overlap (with same item difficulty level) within the

same sub-group. For example, PD03,PD18,PD32,PD22,PD29,PD33 and PD34 are measured at

around -0.2 logit. Similarly, PD06, PD07, PD25, PD15, PD23, PD30 are measured at around -0.3

logit. These items shall be reviewed for redundancy, in order to simplify the questionnaire later.

There are many items below the minimum person measure (-0.62 logit), which are endorsed by

all respondents. These items could be considered for exclusion from the final questionnaire.

Item Fit Statistics

Table 6: Item Fit Statistics

Name Measure Error In.Mnsq In.Zstd Out.Mnsq Out.Zstd PTME


PD06 -0.34 0.14 *2.28 7.66 2.19 5.78 0.40
AS05 2.55 0.14 *1.68 4.81 1.64 4.48 0.44
AS03 -0.32 0.14 *1.65 4.44 1.59 3.29 0.26
AH08 1.56 0.13 *1.58 4.22 1.66 4.54 0.35
PD12 1.53 0.13 *1.55 4.02 1.71 4.83 0.32
PT01 0.06 0.14 1.49 3.52 1.49 3.02 0.41
HA07 1.63 0.13 1.48 3.53 1.46 3.34 0.53
KK09 0.14 0.14 1.44 3.18 1.74 4.35 0.18
PD03 -0.25 0.14 1.43 3.11 1.54 3.07 0.31
AS02 1.34 0.13 1.42 3.13 1.42 3.05 0.39
DM06 1.15 0.13 1.42 3.13 1.4 2.86 0.40
PD20 0.69 0.13 1.39 2.91 1.38 2.6 0.31
PD24 0.45 0.13 1.37 2.72 1.35 2.35 0.37
PD30 -0.34 0.14 1.37 2.7 1.33 1.96 0.40
KK05 0.08 0.14 1.36 2.67 1.34 2.2 0.35
Better fitting items omitted.
11
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

The infit and outfit mean squares for all the items are between 0.50 and 1.50 (Table 2), which are

in the acceptable range (Linacre, 2012), except PD06 (I am not involved in corruption), PD12 (I

have a study time schedule at home), AS05 (I report open burning to the authorities), AS03 (I

write on both side of the paper) and AH08 (I admonish friend who throw garbage everywhere),

are more than 1.50. Therefore, the items need to be revised for modification, deletion or

replacement. Items with a mean square value of more than 1.5 are unproductive for the

construction of measurement. The items require modification of the sentence structure and terms

used.

Item polarity is used as an indicator to clarify whether the items move in one direction as

the construct being measured. As for all the 108 items, the point-measure correlations (PTMEA

Corr.) range from 0.18 to 0.62 (Table 4). A positive index indicates that all the items move in

one direction and measure what the researcher wants to measure (Bond and Fox, 2015). In other

words, the items in the construct of expression do not conflict with the construct to be measured.

Category calibrations

Table 7: Summary of Category Structure


-------------------------------------------------------------------
|CATEGORY OBSERVED|OBSVD SAMPLE|INFIT OUTFIT||STRUCTURE|CATEGORY|
|LABEL SCORE COUNT %|AVRGE EXPECT| MNSQ MNSQ||CALIBRATN| MEASURE|
|-------------------+------------+------------++---------+--------|
| 1 1 398 3| -.27 -.42| 1.14 1.20|| NONE |( -3.04)| 1
| 2 2 2483 19| .35 .40| .93 .95|| -1.83 | -1.00 | 2
| 3 3 5803 45| 1.17 1.16| .93 .89|| -.07 | .97 | 3
| 4 4 4276 33| 2.13 2.13| 1.04 1.03|| 1.90 |( 3.09)| 4
-------------------------------------------------------------------

As for the rating of 1 to 4, the arrangement for the measures of the categories is -0.27, 0.35, 1.17

and 2.13 respectively (Table 5). The threshold values observed from the structure calibration are

in an increasing order. All the categories illustrate clear peaks (Figure 3).
12
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Figure 3: Category Probability Curve

Conclusion

Based on the Wright Map, there are many redundant items with same item difficulty level.

These items shall be further reviewed, to reduce the number of items further. Five item

were found misfittings. Among these, item PD06 (I am not involved in corruption) and

AS05 (I report open burning to the authorities) are highly sensitive items. Respondents

might not given the real responds. Therefore, the items were suggested to be dropped.

There are also many items below the minimum person measure (-0.62 logit), which are

endorsed by all respondents. These items could be considered for exclusion from the final

questionnaire. The reliability and separation indices of the items and are within the
13
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

acceptable range. The 4-point rating scale is suitable to be used for this scale. The stability

of the Students' Moral Behavior Inventory needs to be tested again after the modification of

the misfit items.

References

Andrich, D. (1978). A rating scale formulation for ordered response categories. Psychometrika,

43:561-573.

Bond, T. G., and Fox, C., M. (2015). Appling the Rasch model: Fundamental measurement in the

human sciences. (3rd ed.). New York: Routledge Taylor & Francis publications.

Ezhar T., Jusang B., Mohamad R. H. dan Zamree Y. (2007). "Masalah sosial dan kesedaran sivik

generasi muda daripada keluarga berpendapatan rendah di bandar (social affairs and civic

awareness of youths from low-income families in the city.)". Laporan penyelidikan

(research report). Universiti Putra Malaysia.

Hawley, P. H., & Geldhof, G., J. (2012). Preschoolers’ social dominance, moral cognition and

moral behavior: An evolutionary perspective. Journal of experimental child psychology,

112:18-35. Elsevier publisher.

Ministry of Education. (2000). Secondary school moral education curriculum. Kuala Lumpur:

Curriculum development centre.

Kochanska, G. (1997). Multiple pathways to conscience for children with different temperament:

From toddlerhood to age 5. Developmental psychology, 33:228-240.

Kolhberg, L. (1984). The psychology of moral development: The nature and validity of moral

stages. San Francisco: Harper & Row.


View publication stats

14
STUDENTS' MORAL BEHAVIOUR INVENTORY DEVELOPMENT AND
VALIDATION: A RASCH ANALYSIS

Linacre, J. M., and B. D. Wright. (2012). A user’s guide to WINSTEPS ministeps Rasch model

computer programs. Chicago: Mesa press.

Linacre, J.M. (2003). Dimensionality: Contrasts & variances. Help for winsteps Rasch

measurement software. Retrieved on 19 April 22, 2016 at

http://www.winsteps.com/winman/principalcomponents.htm.

Mumtazah, O., and Nurizan, Y. (2009). Penghayatan nilai-nilai murni dalam kalangan remaja

sekolah (Appreciation of moral values among school youth.). Malaysian journal of social

policy and society, 6:1-20.

Nik Zaharah, N.Y. (2007). Kajian penerapan nilai murni menerusi pengajaran bermodul

(Implementation study of noble values through modular learning). Kuala Lumpur:

Univerisiti Kebangsaan Malaysia.

Sherron, C.T. (2000). Psychometric development of the adaptive leadership competency profile.

Doctorate thesis. University of North Texas.

Vishalache, B. (2012). Dilema kehidupan sebenar dalam pendidikan moral (Real life dilemma in

moral education). Kuala Lumpur: Universiti Malaya.

Willemsem, M. (2012). Exploring a virtue-based moral development approach. Moral

development of dutch high school students. Doctorate dissertation. University Utrecht.

Zawin Aidah. (2010). Sejarah perkembangan nilai, moral dan etika dalam pendidikan (History

of value, moral and ethics in education.). Retrieved on 3 February, 2014 at

http://zawinaida.blogspot.com/2010/11/sejarah-perkembangan-nilai-moral-dan.html.

You might also like