Nama
: Kurnianto
N I M
: 0102512077
Program
: MP,
S2 Kepengawasan
Mata Kuliah
: Bahasa Inggris
Descriptive Research
Descriptive research involves
collecting data in order to test hypotheses or answer questions concerning the
current status of the subject of the study. A descriptive study determines and
reports the way things are. One common type of descriptive research involves
assessing attitudes or opinions toward individuals, organizations, events, or
procedures; pre-election political polls and market research surveys are
examples of th is type of descriptive research. Descriptive data are typically
collected through a questionnaire survey, an interview, or observation.
Descriptive research sounds very
simple; there is considerably more to it, however, than just asking questions
and reporting answers. Since one is generally asking questions that have not
been asked before, instruments usually have to be developed for specific
studies; instrument development requires time and skill. A major problem further
complicating descriptive research is lack of response-failure of subjects to
return questionnaires or attend scheduled interviews. If the response rate is
low, valid conclusions can not be drawn. For example, suppose you are doing a
study to determine attitudes of principals toward research. You send a
questionnaire to 100 principals and ask the question, "Do you usually
cooperate if asked to participate in a research study?" Suppose 40
principals respond and they all answer "yes." Could you then
conclude that principals cooperate? No! Even though all those who responded
said "yes," those 60 who did not respond may never cooperate with
research efforts. After all, they did not cooperate with you! Observational research
also involves complexities that are not readily apparent. Observers must be
trained and forms must be developed 50 that data will be collected objectively
and reliably.
The following are examples of
typical questions investigated by descriptive research studies :
1. How do second·grade
teachers spend their time? Second-grade teachers would be observed for a
period of time and results would probably be presented as percentages, e.g.,
60% of their time is spent lecturing, 20% asking or answering questions, 10%
administering discipline, and 10% performing administrative duties, such as
collecting milk money.
2. How will citizens
of Yortown vote in the next presidential election? A survey of
citizens of Yortown would be taken (questionnaire or interview) , and results
would probably be presented as percentages; e.g., 70% indicate they will vote
for Peter Pure, 20% for George Graft, and 10% are undecided.
3. How do parents
feel about split-shift school days? Parents would be surveyed and results
would probably be presented in terms of the percentages for, against, or
undecided.
Penelitian
Deskriptif
Penelitian deskriptif melibatkan pengumpulan data dalam rangka menguji
hipotesis atau menjawab pertanyaan mengenai status subjek penelitian. Sebuah
studi deskriptif menentukan dan melaporkan situasi yang ada. Salah
satu yang umum dari penelitian deskriptif melibatkan sikap menilai atau
pendapat terhadap individu, organisasi, peristiwa, atau prosedur, jajak
pendapat politik pra-pemiludan survei penelitian pasar adalah contoh dari
adalah jenis penelitian deskriptif. Data deskriptif biasanya dikumpulkan
melalui survei kuesioner, wawancara, atau observasi.
Penelitian deskriptif terdengar sangat sederhana, lebih jauh dari
itu, bukan sekedar bertanya dan melaporkan jawaban. Karena umumnya mengajukan
pertanyaan yang belum ditanyakan sebelumnya, instrumen biasanya harus
dikembangkan untuk studi tertentu, pengembangan instrumen membutuhkan waktu dan
keterampilan. Masalah utama lebih rumit dari penelitian deskriptif adalah
kurangnya respon-kegagalan subyek untuk kembali kuesioner atau menghadiri
wawancara yang dijadwalkan. Jika tingkat respon rendah, tidak dapat ditarik
kesimpulan yang valid . Sebagai contoh, misalkan Anda sedang melakukan
penelitian untuk menentukan sikap kepala sekolah terhadap penelitian. Anda
mengirim kuesioner kepada 100 kepala sekolah dan bertanya, "Apakah Anda
biasanya bekerja sama jika diminta untuk berpartisipasi dalam studi
penelitian?" Misalkan 40 kepala sekolah merespon dan mereka semua menjawab
"ya." Bisakah Anda kemudian menyimpulkan bahwa kepala sekolah bekerja
sama? Tidak ada! Meskipun semua orang yang menanggapi mengatakan
"ya," mereka 60 yang tidak menanggapi mungkin tidak pernah bekerja
sama dengan upaya penelitian. Setelah semua, mereka tidak bekerja sama dengan
Anda! Penelitian observasional juga melibatkan kompleksitas yang tidak nampak.
Pengamat harus dilatih dan bentuk harus dikembangkan 50 bahwa data akan
dikumpulkan secara obyektif dan andal.
Berikut ini
contoh2 pertanyaan tipikal yg investigasi oleh studi penelitian deskriptif:
- Bagaimana guru2 kelas 2 memanfaatkan waktunya? Guru2 kls 2 akan diobservasi
selama satu periode dan hasilnya mungkin akan dipresentasikan sebagai
prosentase misalnya 60% waktunya digunakan mengajar, 20% bertanya atau menjawab
pertanyaan, 10% melaksanakan disiplin administrasi, dan 10% melaksanakan tugas2
administrasi misalnya mengumpulkan uang susu.
- Bagaimana warga Yortown akan memberikan suara pd pemilihan presiden yg akan
dtg? Sebuah survey warga Yortown akan dilakukan (questioner atau interview),
dan hasil2nya mgkn akan disajikan dlm prosentase misalnya 70% menunjukkan
mereka akan memberikan suara utk Peter Pure, 20% utk George Graft, dan 10% tdk
memutuskan.
- Bagaimana para ortu merasakan hari2 sekolah paruh waktu? Orang tua akan
disurvey dan hasilnya mgkn akan disajikan dlm prosentase utk setuju, menolak,
atau tdk memutuskan.
Correlational Research
Correlational research attempts to
determine whether, and to what degree, a relationship exists between two or
more quantifiable variables. The purpose of a correlational study may be to
establish relationship (or lack of it) or to use relationships in making
predictions. Relationship studies typically study a number of variables
believed to be related to a major, complex variable, such as achievement.
Variab!es found not to be highly related are eliminated from further
consideration; variables that are highly related may suggest causal-comparative
or experimental studies to determine if the relationships are causal. For
example, the fact that there is a relationship between self-concept and
achievement does not imply (hat self-concept "causes" achievement or
that achievement "causcs" self-concept. Such a relationship only
indicates that students with higher self-concepts have high er levels of
achievement and students with lower self-concepts have lower levels of
achievement. From the fact that two variables are highly related, one cannot
conclude th at one is the cause of the other; there may be a third factor which
"cau ses" both of the relater' variables. For example, suppose it
were determined th at there is a high degree of relationship between number of
years of schooling and income at age 40 (two quantifiable variables). The
temptation might be to conclude th at if you stay in schoollonger you will make
more money; this conclusion would not necessarily be justified. There might be
a third variable , such as motivation, which "causcs" people to stay
in school and do well in their jobs. The important point to remember is
that correlational research never establishes a cause-effect relationship,
only a relationship.
Regardless of whether a relationship
is a cause-effect relationship, the existence of a high relationship perrnits
prediction. For example, high school grades and college grades are highly
related; students who have high GPAs in high school tend to have high GPAs in
college, and students who have low GPAs in high school tend to have low GPAs in
college. Therefore, high school GPAs can Le, and are, used to predict GPA in
college. The degree of relationship between two variables is generally
expressed as a correlation coefficient, which is a number between .00 and 1.00.
Two variables that are not related will produce a coefficient near .00; two
variables that are highly related will produce a coefficient near 1.00. Since
very few relationships are perfect, prediction is rarely perfect. However, for
many decisions, predictions based on known relationships are very useful.
The following are examples of
typical correlational studies:
1.
The relationship between intelligence and creativity. Scores on an
intelligence test and on a creativity test would be acquired from each member
of a given group. The two sets of scores would be correlated and the resulting
coefficient would indicate the degree of relationship.
2. The
relationship between anxiety and achievment. Scores on an anxiety scale and
on an achievement test would be acquired from each member of a group. The two
sets of scores would be correlated and the resulting coefficient would indicate
the degree of relationship.
3. Use of an
aptitude test to predict success in an algebra course. Scores on an
algebra aptitude test would be correlated with ultimate success in algebra as
measured by final exarn scores, for example. If the resulting coefficient were
high, the aptitude test would be considered a good predictor.
Terjemahan
Penelitian
Korelasi
Berikut ini
contoh2 tipikal studi korelasi:
1.
Hubungan antara intelegensia dg kreativitas. Skor atas uji intelegensia dan pd
uji kreativitas akan diperoleh dr masing anggota kelompok yg ada. 2 kelompok
skor akan dikorelasikan dan koefesien hasilnya akan menunjukkan tingkat
keterkaitannya.
2.
Hubungan antara kecemasan dg prestasi. Skor2 pd skala kecemasan dan pd uji
prestasi akan diperoleh dr masing2 anggota kelompok. 2 kelompok skor akan
dikorelasikan dan koefesien hasilnya akan menunjukkan tingkat keterkaitannya.
3.
Penggunaan tes bakat utk memprediksikan kesuksesan di sebuah kursus aljabar.
Skor2 pd uji bakat aljabar akan dikorelasikan dg kesuksesan akhir dlm aljabar
krn diukur dg skor ujian akhir, Misalnya. Jika koefesien hasilnya tinggi, maka
uji bakat akan dianggap prediktor yg baik.
Causal-Comparative and Experimental
Research
While causal-comparative and
expenrnantal research represent distinctly different methods, they can best be
understood through comparison and contrast. Both attempt to establish
cause-effect relationships; both involve group comparisons. The major difference
between them is that in experimental research the alleged "cause" is
manipulated, and in causal-comparative research it is not. In experimental
research, the alleged "cause, " the activity or characteristic
believed to make a difference, is referred to as a treatment; the more general
term for "cause" is independent variable. The difference, or
"affect," which is determined to occur or not occur is referred to as
the dependent variable. Dependent on what? Dependent on the independent
variable. Thus, a study which investigates a cause-effect relationship
investigates the effect of an independent variable on a dependent variable.
ln an experimental study the
researcher manipulates at least one independent variable and observes the
effect on one or more dependent variables. In other words, the researcher
determines "who gets what," which group of subjects will get whlch
treatment; the groups are generally referred to as experimental and control
groups. Manipulation of the independent variable is the one single
characteristic that differentiates experimental research from other methods.
Ideally, in experimental research the groups to be studied are randomly formed
before the experiment, a procedure not involved in the other methods of
research. The essence of experimentation is control. The researcher strives to
insure th at the experiences of the groups are as equal as possible on all important
variables except, of course, the independent variable. If, at the end of same
period of time, groups differ in performance on the dependent variable, the difference
can be attributed to the independent variable. Because of the direct manipulation
and control of variables, experimental research is the only type of research
that can truly establish cause-effect relationships.
The folIowing are exarnples of
typical experimental studies:
1.
The comparative effectiveness of programmed instruction versus traditional
instruction on computational skill. The independent variable, or
cause, is type of instruction (programmecLversus traditional); the dependent
variable, or effect, is computational skill. Two groups (preferably randomly
formed) would be exposed to essentially the same experiences, except for method
of instruction. After same period of time, their computational skill would be
compared.
2. The effect of
self-paced instruction on self-concept. The independent variable, or
cause, is pacing (self-pacing versus teacher pacing); the dependent variable,
or effect, is self-concept. Two groups (preferably randomly formed) would be
exposed to essentially the same experiences, except for the pacing of
instruction. After some period of time, their self-concepts would be compared.
3.
The effect of positive reinforcement on attitude toward school. The
independent variable, or cause, is type of reinforcement (e.g., positive versus
negative, or positive versus none) ; the dependent variable , or effect, is
attitude toward school. Two groups (preferably randomly formed) would be
exposed to essentially the same experiences, except for the type of
reinforcement received. After same period of time, their attitudes toward
school would be compared.
In a
causal-comparative study the independent variable, or "cause”: is not
manipulated; it has already occurred. Independent variables in
causal-comparative studies are variables which can not be manipulated (e.g.,
sex, male-female), should not be rnanipulated (e.g., brain damage) , or simply
are not manipulated, but could be (e.g., method of instruction). In
causal-comparative research, groups are also compared on some dependent
variable; these groups, however, are different on some variable before the study
begins. Perhaps one group possesses a characteristic and one does not, or perhaps
each group is a mem ber of a different socio economic level. In any event, the
diiference between the groups (the independent variable) is not, was not, or
could not be determined by the researcher. Further, since the independent
variable has already occurred, the same kinds of controIs cannot be exercised
as in an experimental study. Due to the lack of manipulation and control,
cause-effect relationships established are at best tenuous and tentative. On
the positive side, causal-comparative studies are less expensive and take much
less time to conduct. Further, apparent cause-effect relationships may lead to
experimental studies deslcned to confirm or disconfirm the findings.
Also , there are a number of important variables which simply cannot be
manipulated. Studies designed to investigate the effects of a broken home,
intelligence, or sex on achievement must be causal-comparative, as none of
these variables can be manipulated. The folIowing are examples of typical
causal-comparative studies:
1. The effect of kindergarten
attendance on achievement at the end of the first grade.
The independent variable, or cause,
is kindergarten attendance (students attended kindergarten or they did not);
the dependent variable, or effect, isachievement at the end of the first grade.
Two groups of first graders would be identified-one group who had attended
kindergarten and one group who had not. The achievement of the two groups
would be compared.
2. The effect
of having awarking mother on school absenteeism. The independent
variable, or cause, is the employment status of the mother (the rnother works
or does not work); the dependent variable, or effect, is absenteeism, or number
of days absent. Two groups of students would be identified-one group who had
working mothers and one group who did not. The absenteeism of the two groups
would be compared.
3. The effect
of sex on algebra achievement. The independent variable, or cause,
is sex (male versus female); the dependent variable, or effect, is algebra
achievement. The achievement of males would be compared to the achievement of
females.
Terjemahan
Penelitian
Kausal-Komparatif dan Eksperimental
Berikut ini
adalah contoh2 tipikal studi eksperimental:
1.
Keefektifan pengajaran terprogram dibandingkan dg pengajaran tradisional atas
ketrampilan berhitung. Variabel bebas atau sebab adalah jenis pengajaran
(terprogram vs tradisional); variabel terikat atau akibat adalah ketrampilan
berhitung. Dua kelompok (lebih dipilih terbentuk acak) akan dibuka pd dasarnya
mjd pengalaman yg sama, kecuali utk metode pengajaran. Setelah beberapa waktu
ketrampilan berhitung mereka akan dibandingkan.
2.
Akibat pengajaran kemauan sendiri pd konsep pribadi. Variabel bebas atau sebab
adalah kemauan sendiri/belajar bebas (kemauan sendiri vs langkah guru);
variabel terikat atau akibat adalah konsep pribadi. 2 kelompok (diambil secara
acak) akan dibuka/dipilih secara jelas sbg pengalaman yg sama, kecuali utk
langkah/laju pengajaran. Setelah beberapa waktu konsep2 pribadi mereka akan
dibandingkan.
3.
Akibat penguatan positif atas sikap thd sekolah. Variabel bebas atau sebab
adalah jenis penguatan (misalnya positif vs negative, atau positif vs tdk ada);
variabel terikat atau akibat adalah sikap thd sekolah. 2 kelompok( dipilih
secara acak) akan dibuka secara jelas mjd pengalaman yg sama, kecuali utk tipe
penguatan yg diterima. Setelah beberapa waktu sikap mereka thd sekolah akan
dibandingkan.
Berikut ini
adalah contoh2 tipikal studi kausal-komparatif:
1. Akibat
kehadiran siswa TK pd prestasi di akhir kelas pertama. Variabel bebas atau
sebab adalah kehadiran siswa TK (siswa TK yg dihadirkan atau yg tdk
dihadirkan); variabel terikat atau akibat adalah prestasi di akhir kelas
pertama. 2 kelompok yg diberi nilai pertama akan diidentifikasi–satu kelompok
yg dating di TK dan satu kelompok yg tdk hadir. Prestasi 2 kelompok akan
dibandingkan.
2. Akibat
memiliki ibu yg bekerja pd ketidakhadiran di sekolah. Variabel bebas atau sebab
adalah status ibu ( ibu bekerja atau tdk); variabel terikat atau akibat adalah
ketidakhadiran atau jumlah hari2 absen/tdk hdr. 2 kelompok siswa akan
diidentifikasi–satu kelompok yg memiliki ibu yg bekerja dg satu kelompok yg
ibunya tdk bekerja. ketidakhadiran 2 kelompok akan dibandingkan.
3. Akibat
jenis kelamin pd prestasi matematika. Variabel bebas atau sebab adalah jenis
kelamin (laki2 vs perempuan); variabel terikat atau akibat adalah prestasi
matematika. Prestasi laki2 akan dibandingkan dg prestasi perempuan.
Guidelines for Classification
Which of the five methods is most
appropriate for a given study depends upon the way in which the problem is
defined. The same general problem can often be investigated using several of
the methods. Research in a given area is often scquential; preliminary
dcsctiptive and/or correlational studies may be conducted followed by causalcomparative
and/or experimental studies, if such seem warranted. Asan example, let us look
at anxiety and achievement. The folIowing studies might be conducted:
a .
Descriptive: A survey of teachers to determine how and to what degree they
believe anxiety affects achievement.
2.
Correlational: A study to determine the relationship between scores on an
anxiety scale and scores on an achievement measure.
3.
Causal-comparative: A study to compare the achievement of a group of students
classified as high-anxious and a group classified as low-anxious.
4.
Experimental: A study to compare the achievement of two groups-one group taught
in an anxiety-producing environment and one group taught in an anxietyreducing
environment."
When analyzing a study in order to
determine the method represented, one approach is to ask yourself the
folIowing series of questions. First, Was the researcher attempting to
establish a cause-effect relationship? If yes, the research is either causalcomparative
or experimental. The next question is, Was the alleged cause, or independent
variable, manipulated by the researcher? Did the researcher control who got
what and what they got? If yes, the research is experimental; if no, the
research is causal-comparative. If the answer to the very first question is no,
the next question should be, Was the researcher attempting to establish a
relationship or use a relationship for prediction? If yes, the research is
correlational. If no, the research is either descriptive or historical, and
you should have no difficulty discriminating between the two (see Figure 1.1).
The folIowing examples should further clarify the differences among the
methods:
1.
Teacher attitudes toward unions. Probably descriptive. The
study is determining the current attitudes of teachers. Data are probably
collected through use of a questionnaire or an interview.
2.
Effect of socioeconomic status (SES) on self-concept. Probably
causal-comparalive. The effect of SES on self-concept is being investigated.
The independent variable , socioeconomic status, cannot be manipulated.
3.
Comparison of large-group versus small-group instruction on achievement.
Probably experimental. The effecl of size of group on achievement is being
investigated. The independent variable, group size, can be manipulated by the
researcher.
Terjemahan
Petunjuk2
Klasifikasi
Manakah dr 5
metode yg plg tepat utk studi yg ada tergantung pd cara dimana masalah
dibatasi. Masalah umum yg sama sering bisa diinvestigasi dg menggunakan
beberapa metode. Penelitian dlm wilayah yg ada sering runtut; deskriptif
preliminer dan atau studi korelasi bisa dilakukan yg diikuti dg
Kausal-komparatif dan atau studi eksperimental, jika semacam itu kelihatannya
terjamin. Sebagai contoh, marilah kita lihat kecemasan dg prestasi. Studi2
berikut ini mgkn bisa dilakukan:
1. Deskriptif:
sebuah survey guru utk menentukan bagaimana dan pd tingkat apa mereka percaya
kecemasan berakibat pd prestasi.
2. Korelasi:
sebuah studi utk menentukan hubungan antara skor pd skala kecemasan dg skor pd
ukuran prestasi.
3. Kausal-komparatif:
sebuah studi utk membandingkan prestasi sebuah kelompok siswa dikelompokkan yg
memiliki kecemasan tinggi sama dg kelompok yg memiliki kecemasan yg
rendah.
4. Eksperimental:
sebuah studi utk membandingkan prestasi 2 kelompok–satu kelompok dibelajarkan
dlm lingkungan yg menghasilkan kecemasan dg satu kelompok yg dibelajarkan dlm
lingkungan yg mengurangi kecemasan.
Berikut
contoh2 seharusnya lebih jauh mengklarifikasikan perbedaan antar metode:
1. Sikap
guru thd persatuan. Mungkin deskriptif. Studi menentukan sikap guru sekarang.
Data mungkin dikumpulkan melalui penggunaan questioner atau interview.
2. Akibat status
sosioekonomik (SES) pd konsep pribadi. Mgkn kausal-komparatif. Akibat SES pd
konsep pribadi sdg diteliti. Variabel bebas, status sosioekonomik, tdk bisa
dimanipulasi.
3. Perbandingan
pengajaran kelompok besar vs kelompok kecil thd prestasi. Mgkn eksperimental.
Akibat ukuran kelompok pd prestasi sdg diteliti. Variabel bebas, ukuran
kelompok, bisa dimanipulasi oleh peneliti.
