Research
review
By
: Dr. Suwandi Mpd (Lecturer of IKIP PGRI and Diponegoro University Semarang)
What is research?
Research is a systematic attempt to provide answers to
questions
Characteristics of a research
- Research is directed
toward the solution of a problem
- Research emphasizes
the development of generalization, principles, or theories that will be
helpful in predicting future occurances
- Research involves
gathering new data from primary or
first hand sources or using existing data for a new purpose
- Research demands
accurate observation and desrciption
- Research is based upon
observable experience or emphirical evidance
Selecting a problem
A problem statement must have the following
characteristics:
- It should ask about a
relationship between two or more variables
- It should be stated
clearly and unambiguously, usually in question form
- It shoud be testable
by empirical methods; thatis, it should be possible to collect data to
answer the questions asked
- It should not
represent a moral or ethical position
The problem is stated in question form
- What is the relationship
between i.q. and achievement?
- Do students learn more
from a directive teacher or a nondirective teacher?
- Is there a
relationship between racial background and dropout rate?
- What is the
relationship between rote learning ability and socio-economic status?
Formulating hypotheses
The next step after the selection of the problem is to state a hypothesis. A
hypotesis, a suggested answer to a problem, has the following
characteristics:
- It should
conjucture upon a relationship
between two or more variables.
- It should be stated
clearly and unambiguously in the form
of a derclarative sentence
- It should be testable;
that is, it should be possible to restate it in an operational form that
can then be evaluated based on data
The hypotheses of the stated problems
- I.q. and achievement
are positively related
- Directive teachers are
more effective than nondirective
teachers
- The dropout rate is
higher for black students than for white students
- Error rate in a rote
learning task and socioeconomic status are inversely related; that is,
middle-class youngsters make fewer errors in a rote learning task than
lower-class youngster
Experimental research
Experiments are carried out in order to explore the
strength of relationship between variables
Variables
Variables are the conditions or characteristics that
the experimenter manipulates, controls, observes
Independent variables
Independent variables are the conditions or
characteristics that the experimenter manipulates in his attempt to ascertain
their relationship to observed phenomena
Dependent variables
Dependent variables are the conditions or
characteristics that appear, disappear or change as the experimenter
introduces, removes, or changes independent variables
Independent variables
Independent variables can be a particular teaching
method, a type of teaching material, a reward, a period of exposure to a
particular condition
Dependent variables
Dependent variables can be a test score, a number of
errors, or measured speed in performing a task.
Thus, the dependent variables are the measured changes
in pupil performance attributable to the influence of the independent variables
Organismic or attribute variables
} Organismic variables are those characteristics that
can not be altered by the experimenter, such age, sex, or race.
} The question whether eight-year old girls show greater
reading achievement than eight-year old boys provide an axample of sex, as an
organismic variable
Intervening variables
} Intervening variables are variables that intervene
between the cause and effect. The variables can not be controlled or measured ,
such as anxiety, fatigue, motivation.
Experimental studies
In an experimental study, the investigator deliberately manipulates the independent
variable in order to test the effect on
the dependent variable
Experimental design
·
Experimental
design is the blue print of the procedures that enable the researcher to test
his hypotheses by reaching valid conclusions about relationships between
independent and dependent variables
·
Symbols used in
exp design
r= random
selection of subjects
x=
experimental variable
c= control
variable
o=
observation or test
Three kinds of experimental design
- Pre-experimental
design
- Quasi-experimental
design
- True experimental design
} Pre-experimental design
- The one-shot case
study
x o
- The one-group,
pretest-posttest design
o1
x o2
- The static-group
comparison design
x o1
x o2
} Quasi-experimental designs
- The nonequivalent,
pretest-postest design
o1 x
o2
o3
o4
- The
equivalent-materials, single-group, pretest-posttest design
ma o1
x o2; mb
o3 c o4
} True experimental design
1. The posttest-only, equivalent-group design
r x
o1
r o2
2.
Pre-test-posttest equivalent-groups design
r o1
x o2
r o3
o4
Descriptive studies: describes and
interprets what is.
It covers:
} Assessment: describes the status of a phenomenon at a particular
time
} Evaluation:
school surveys are usually evaluation studies; educational products and
programs are examined to determine their effectiveness in meeting
objectiveness, often with recommendations for constructive action
} Descriptive research: known as nonexperimental
research, deals with the relationship between variables, the testing of
hypotheses, and the development of generalization, principles, or theories that
universal validity.
Assessment and evaluation studies
·
Assessment:
- The survey
- The case study
- Social survey
- Public-opinion survey
·
Evaluation:
- National assessment of
educational progress
- International
assessment
Descriptive research
descriptive research seeks to find the
answers to questions through the analysis of variable relationships. For exmpl,
excessive speed were associated with a high proportion of fatal accidents.
Contrasting pre-experiment, quasi-experiment, true experiment
} Pre-experiment: may have pre- and posttreatment tests,
but lacks a control group
} Quasi-experiment: has both pre- and posttest and
experimental and control groups, but no random assignment of subjects
} True experiment: has both pre- and posttests,
experimental and control groups, and random assignment of subjects
Correlational studies
Correlational studies are usually intended to answer
three questions about two variables or two sets of data.
} First: is there a relationship between the two
variables (or sets of data)?
} Second: what is the direction of the relationship?
} Third: what is the magnitude?
Correlation
Do high values of variable x tend to go together with
high values of variable y. Or do high values of x go with low values of y. Or
is there some complex relationship between x and y, or perhaps no
relationship at all.
An
example of a correlational study
} Table 1 the language test scores and their squares and
products
|
Scores on
|
||||
|
Comprehension
(x)
|
Translation
(y)
|
X²
|
Y²
|
Xy
|
|
17
|
15
|
289
|
255
|
255
|
|
13
|
13
|
169
|
169
|
169
|
|
12
|
8
|
144
|
64
|
96
|
|
14
|
17
|
196
|
289
|
238
|
|
15
|
16
|
225
|
256
|
240
|
|
8
|
9
|
64
|
81
|
72
|
|
9
|
14
|
81
|
196
|
126
|
|
13
|
10
|
169
|
100
|
130
|
|
11
|
16
|
121
|
256
|
176
|
|
14
|
13
|
196
|
169
|
182
|
|
12
|
14
|
144
|
196
|
168
|
continued….
|
Scores on
|
||||
|
Comprehension
(x)
|
Translation
(y)
|
X²
|
Y²
|
Xy
|
|
16
|
17
|
256
|
289
|
272
|
|
10
|
9
|
100
|
81
|
90
|
|
18
|
16
|
324
|
256
|
288
|
|
15
|
19
|
225
|
361
|
285
|
|
16
|
16
|
256
|
256
|
256
|
|
12
|
17
|
144
|
289
|
204
|
|
16
|
12
|
256
|
144
|
192
|
|
10
|
14
|
100
|
196
|
140
|
|
9
|
8
|
81
|
64
|
72
|
|
Sx = 260
|
Sy = 273
|
Sx²= 3540
|
Sy² = 3937
|
Sxy= 3651
|
Correlation measurement
Pearson product-moment
correlation coeficient
R = nSxy-Sx Sy
√{nSx² -(Sx)²}{nSy²-Sy²}
R = (20
x 3651) – (260 x 273)
√ {(20
x 3540) - 260²}{(20 x 3937) - 273²}
73020 - 70980
=
√(70800 – 67600)(78740 –
74529)
= 2040 2040
√3200 x 4211 = 3670.86 = 0.56
Reporting correlation studies
In reporting correlations one has to report the test
used, the cooeficient, the direction of
the correlation (positive or negative ),
and the significance level (if considered appropriate). For expl: using the
pearson-product moment correlation, a statistically significant correlation was
found between students’ attendance at school and their exam performance (r=
0.87, ρ= 0.035). Those students who attended school the most
tended to have the best exam performance and those who attended the least
tended to have lowest exam performance
Suggested research topics
An examination of english objective test items from
the uan 2007 for smp
Reading comprehension exercises with accompanying
tests for year seven of smp
Suggestions for games and activities for assisting the
young learner of english in elementary school year 4.
Attitudes, motivation and socio cultural variables in
english competence: a questionaire
Some suggested techniques and activities for
developing fluency in spoken english of sma students
Games and contests in english language teaching
An analysis of the errors made by the english students
of ikip pgri
The appropriateness of reading materials for
polytechnic students
High frequency errors made by the third semester
students of the english department of the faculty of language education ikip
pgri semarang
No comments:
Post a Comment