Split-Halves Method
Uploaded by
ScribdTranslationsSplit-Halves Method
Uploaded by
ScribdTranslationsFACULTY OF SCIENCES
SCHOOL OF CHEMICAL SCIENCES
ENGINEER IN ENVIRONMENTAL BIOTECHNOLOGY
U.I.C. EPISTEMOLOGY AND METHODOLOGY
Split-halves method
AUTHORS:
Chimbo Valeria Pomagualli Javier
Garcés Diego Benjamín Toaquiza
Alexis Maldonado Óscar Umajinga
TEACHER: Eng. Silvana Paola Ocaña
JUNE 22, 2020
RIOBAMBA - ECUADOR
I. OBJECTIVES
1
The objective of this research work is to understand methods for
estimate reliability.
to know the split-halves method and how to calculate it through
correlation coefficient between the scores of the halves of the instrument.
Establish the importance of determining the reliability of the instruments.
measurement in a research.
II. THEORETICAL FRAMEWORK
Importance of determining the reliability of measurement instruments in a
research
Reliability is understood as a property of the test scores that it denotes.
the proportion of true variance and is linked to measurement error. Therefore, to
greater reliability, lower measurement error. It is calculated and evaluated for everything
measurement instrument used, or if several instruments were administered,
determine for each of them. Likewise, it is common for the instrument to contain
various scales for different variables or dimensions, then the reliability is
establishes for each scale and for the total of scales (if they can be summed, if they are additive).
(Sampieri, 2014)
There are various procedures to calculate the reliability of an instrument,
they all use formulas that produce reliability coefficients that can vary between
zero and one, where a coefficient of zero means no reliability and one represents
a maximum of reliability. The closer the coefficient is to zero (0), the greater the error
There will be in the measurement The researcher calculates its value, declares it, and submits it to scrutiny of
the users of the study or other researchers, specifying the method used, how much
more information is provided about reliability, the reader forms an idea
clearer about its calculation and the conditions under which it was demonstrated. (Sampieri, 2014)
In that sense, reliability and validity are important properties of being
reported because it allows the reader to know the level of precision and evidence of the
instruments used, which will lead to coherent conclusions in the study. (Sampieri,
2014)
Method of Split Halves
The split-half method requires only one application of the measurement.
Specifically, the whole set of items (or components) is divided into two halves or
It can be divided by taking the even or odd items, and the scores or results from
both are compared[ CITATION And14 9226 ]The correlation between the two is obtained.
resulting scores and the Spearman-Brown formula is used to estimate the
reliability of the complete instrument.[ CITATION Bar07 l 9226 ]
The Spearman-Brown prediction formula, also known as the prophecy formula
Spearman-Brown is a formula that relates psychometric reliability to test
2
length and used by psychometrists to predict the reliability of a test after
change the length of the essay[ CITATION Bar11 9226 ]
The formula is as follows:
Where:
Rxx'Reliability coefficient when its length has been doubled.
Rxx’Reliability coefficient of the two halves.[ CITATION Cha12 9226 ]
If the instrument is reliable, the scores of both halves must be strongly
correlated. An individual with a low score in one half will tend to also have
a low score in the other half[ CITATION And14 9226 ].
Reliability varies according to the number of items included in the instrument.
measurement. The more items, the higher the reliability increases (of course, as long as they refer to the
same variable). This makes sense, let's look at a daily example: If you want to
to test how reliable or consistent a friend's loyalty is towards us,
the more tests we apply, the greater its reliability will be. Of course, too many
items will cause fatigue in the respondent.
3
EXAMPLE
A pilot test was conducted in which 5 surveys were administered to the oven owners.
bricklayers. The purpose of this pilot test was to determine if these people understood
correctly the questionnaire and if it was suitable for the type of information being
required for the research. A reliability calculation was also performed using the
split-halves method, which consists of producing equal results to
apply the instrument to the same subject or object. The method of split halves requires a
application of measurement to the entire set of components; it is divided into two
halves and a comparison of the results of both is made. If the result of both
Half is similar, it is concluded that the instrument is reliable. (Arevalo et al., 2016)
4
Based on these results, it was observed that the percentages of the responses were very
similar. In the first half, the result was: YES 18.4% and NO 81.6%. In the second
mitad: SI 16.3% y NO 83.7%.
Since the result of the pilot test and reliability was satisfactory, it was decided
include the information from these surveys in the total sample number. And with the result
From the reliability calculation, it could be stated that the instrument was reliable.
III. CONCLUSIONS
Research instruments are resources used to record information.
what is intended to be obtained or measured.
If the instrument is reliable, the scores of the two halves must be very
correlated.
The most commonly used reliability test is Cronbach's alpha, even in its use.
through SPSS for dichotomous items, but it is more appropriate to use the method of
halves due to the freedom to design the number of items in the instrument.
IV. BIBLIOGRAPHY
Andrade, J. (2014). Study Techniques. Retrieved on June 20, 2020, from
Unable to access the content of the provided URL.
split-halves method. In
change, the method of results of both
are compared.
Barraza, A. (2007). Pedagogical University of Durango. Retrieved on June 20th.
2020, from Dialnet-How to Evaluate a Reliability Coefficient-2292993.pdf
Barrios, M. (2011). Open University of Catalonia. Retrieved on June 20, 2020.
from http://openaccess.uoc.edu/webapps/o2/bitstream/10609/69325/3/Psicometr
Day_Module_2_Reliability.pdf
Chacon, S. (2012). Evaluation of the measurement instrument. Retrieved on June 20
2020, from https://personal.us.es/vmanzano/docencia/psicometria/ppt/tema5.2.pdf
Arevalo, D., & Padilla, C. (2016). Measurement of Learning Reliability
RStudio Program Using Cronbach's Alpha Measurement Reliability RStudio
Learning Program using Cronbach's alpha. ResearchGate, 2(January), 1–9.
5
Sampieri, R. H. (2014). Research methodology.
