mixed methode research.docx

Upload: hasril-ibnu-hasan

Post on 02-Mar-2018

251 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    1/42

    ResearchDesign diualitative/ QuantitativeMixedMethoDsM ichael R. h

    aRwell

    University of Minnesota

    Tdia editor dari buku ini menegaskan bahwa Permintaan pendidikan terhubung lemah untuk

    mempelajari ide-ide yang menjanjikan. Itu Implikasinya adalah bahwa penelitian pendidikanbanyak memiliki sebuah "mur dan baut" karakter yang membatasi kreatif Generasi masalah

    atau pemecahan masalah dan enggan mengambil risiko. onsekuensi dari ini praktek peluang

    untuk memajukan terjawab kami memahami masalah pendidikan ofimportant dan

    solutions.!his adalah pernyataan yang berani tapi itu adalah salah satu yang didukung oleh

    pemeriksaan diterbitkan dan studi pendidikan yang tidak dipublikasikan. #eorang pemain

    kunci dalam proses ini merupakan penelitian studi desain. $alam banyak studi pendidikan

    penelitian desain mencerminkan "cookie cutter" pendekatan di mana desain yang sama

    berulang-ulang dan sempit diterapkan dalam cara yang dapat membatasi apa yang dipelajari

    dan bagaimana ide studied.Research berbasis yang tidak bisa dengan mudah dipetakan ke

    sejumlah kecil desain diterima penyandang dana dan jurnal profesional kemungkinan akan

    ditinggalkan atau dimodifikasi agar sesuai banyak digunakan desain penelitian.Penegasan dari bab ini adalah bahwa penelitian ini ide menjanjikan dalam pendidikan akan

    baik dilayani oleh peningkatan penggunaan desain penelitian yang menarik pada beberapa

    tradisi in%uiry.In beberapa kasus desain ini saat ini tersedia dan tantangannya adalah untuk

    lebih peneliti untuk mempekerjakan mereka &atau terus mempekerjakan mereka' dalam ketat

    manner.In kasus lain desain baru akan perlu untuk dikembangkan. (fek bersih adalah bahwa

    katalog penelitian desain yang tersedia untuk pendidikan peneliti adalah memperluas secara

    dramatis. Ini Perubahan tidak mengancam atau melemahkan nilai desain yang saat ini banyak

    digunakan tetapi melainkan meningkatkan kolam desain penelitian tersedia untuk

    mendukung penyelidikan ketat.

    )ab ini mengeksplorasi peran penelitian desain dalam studi ide menjanjikan di pendidikan

    tion dengan asumsi dua kondisi yang present.*irst adalah bahwa ide yang dikejar bernilai

    mengejar yaitu sebuah ide yang cukup dapat menyebabkan dan dukungan pembentukan

    penting pendidikan pertanyaan dan identifikasi solusi untuk masalah pendidikan penting.

    edua adalah bahwa setiap tahap proses penelitian terkait dengan mempelajari ide

    menjanjikan adalah ketat. +on ini ditions menyederhanakan pemeriksaan peran desain

    penelitian dalam studi ide menjanjikan.

    arakteristik tiga penelitian methodologies &metode kualitatif metode kuantitatif metode

    campuran' dan peran mereka dalam mempelajari ide-ide

    Halaman 2

    Sectionaku aku aku.HaippoRtunities danchallenges diDesigning danconDucting

    sayanQuiRy

    diyakini bernilai belajar dijelaskan. saya menggunakan dua studi untuk menggambarkan

    kekuatan dalam penelitian desain serta kesempatan untuk meningkatkan hasil dan

    memberikan perhatian yang lebih besar untuk campuran methods karena kebaruan relatif

    mereka dan materi bagi esensial. !iga cara yang peneliti dapat meningkatkan kemampuan

    desain penelitian untuk lebih mendukung #tudi ide menjanjikan dalam studi pendidikan

    dijelaskan. #aya menyimpulkan dengan menyatakan bahwa dicampur metode menawarkan

    jalan sangat menjanjikan menuju menggunakan desain penelitian dengan cara yang

    mendukung rigor Permintaan ous.

    ResearchDesign

    $alam penelitian pendidikan biasanya mungkin &dan tentu saja populer' untukmengkarakterisasi penelitian metodologi penelitian sebagai kualitatif, sebagai kuantitatif

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    2/42

    tie, atau sebagai melibatkan kualitatif dan kuantitas produk metode tatie dalam hal ini

    biasanya disebut metode sebagai campuran. Penelitian jangka desainsecara luas digunakan

    dalam pendidikan namun mengambil arti yang berbeda dalam studi yang berbeda. &Itu istilah

    metode penelitian dan desainpenelitianakan digunakan secara bergantian dalam bab ini.'

    ntuk-contoh ple dalam sebuah penelitian desain penelitian mungkin mencerminkan

    seluruh proses penelitian dari pembuatan konsep masalah untuk tinjauan literatur penelitian-pertanyaan tions metode dan kesimpulan sedangkan di studi lain desain penelitian hanya

    mengacu pada metodologi penelitian &misalnya pengumpulan data dan analisa'. Mungkin

    tidak mengherankan ada ariasi yang tion dalam dan di antara metodologi bagaimana $esain

    penelitian didefinisikan. /amun ini ariasi yang tion tidak mempengaruhi pemeriksaan peran

    desain penelitian dalam mempromosikan studi ketat menjanjikan ide dan dengan demikian

    definisi tunggal penelitian desain tidak diadopsi dalam bab ini. #aya berasumsi bahwa

    pertanyaan penelitian yang mengemudi kekuatan di balik pilihan desain penelitian dan setiap

    perubahan yang dibuat untuk elemen desain sebagai studi terungkap.

    Mengidentifikasi desain penelitian studi adalah important karena memberikan informasi

    tentang fitur kunci dari penelitian yang dapat berbeda untuk kualitatif kuantitatif danmetode campuran. /amun salah satu fitur umum di seluruh penelitian desain adalah bahwa

    pada satu atau lebih poin dalam proses penelitian data dikumpulkan &nomor kata-kata gerak

    tubuh dll' meskipun dalam cara yang berbeda dan untuk tujuan yang berbeda. $engan

    demikian studi kualitatif adalah antara lain penelitian yang mengumpulkan dan

    menganalisis data kualitatif, studi kuantitatif adalah antara lain penelitian yang

    mengumpulkan dan menganalisis data kuantitatif, dan sebagainya.

    +rotty &0112' menggambarkan empat fitur kunci dipertimbangkan dalam desain penelitian3

    epistemologi yang menginformasikan penelitian filosofis sikap yang mendasari metodologi

    yang bersangkutan &misalnya pasca-positiisme konstruktiisme pragma- tism adokasi 4

    partisipatif, melihat Morgan 5667' metodologi itu sendiri dan teknik-teknik dan prosedur

    yang digunakan dalam desain penelitian untuk mengumpulkan data. *itur-fitur ini

    menginformasikan deskripsi penelitian desain di bawah ini.

    Metode Penelitian Kualitati

    Metode penelitian kualitatif fokus pada disco kenai dan memahami pengalaman per-

    P8/$8/G dan pikiran peserta-yang penelitian kualitatif mengeksplorasi makna tujuan

    atau kenyataan &9iatt 012:'. $engan kata lain penelitian kualitatif adalah kegiatan yang

    terletak menempatkan pengamat di dunia. !erdiri dari dari satu set interpretatif praktek

    materi yang membuat dunia terlihat. Praktik-praktik ini mengubah dunia. Mereka mengubah

    dunia menjadi sebuah serangkaian representasi termasuk catatan lapangan wawancara

    percakapan foto rekaman dan memo untuk diri sendiri. Pada tingkat ini penelitian kualitatif

    melibatkan penafsiran pendekatan naturalistik kepada dunia. Ini berarti bahwa penelitikualitatif mempelajari hal-hal dalam pengaturan alam mereka mencoba untuk membuat rasa

    atau menafsirkan fenomena dalam hal makna orang membawa kepada mereka. &$en;in hal. ?' Pusat untuk penyelidikan ini adalah adanya mul- tiple "kebenaran"

    yang dibangun secara sosial &=incoln < Guba 012>'. Penelitian kualitatif biasanya

    digambarkan sebagai memungkinkan eksplorasi rinci dari topik yang menarik di mana

    informasi dikumpulkan oleh peneliti melalui studi kasus etnografi bekerja wawancara dan

    sebagainya. Melekat dalam ini Pendekatan adalah deskripsi dari interaksi antara peserta dan

    peneliti di naturalistik pengaturan dengan beberapa batasan sehingga fleksibel dan proses

    penelitian terbuka. Ini unik interaksi tions menyiratkan bahwa hasil yang berbeda dapat

    diperoleh dari peserta yang sama tergantung pada siapa peneliti adalah karena hasil

    diciptakan oleh peserta dan peneliti dalam situasi tertentuHalaman 3

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    3/42

    !a" #$. Desain Penelitian Kualitati di / Kuantitati / Mixed Metode&pp. ?1-@6'. $engan

    demikian peniruan dan generali;ability umumnya tidak tujuan penelitian kualitatif.

    Metode penelitian kualitatif juga dijelaskan sebagai induktif dalam arti bahwa peneliti dapat

    membangun teori atau hipotesis penjelasandan konseptualisasi dari rincian yang diberikan

    oleh peserta. !ertanam dalam pendekatan ini adalah perspektif bahwa para peneliti tidak

    dapat menyisihkan mereka pengalaman persepsi dan bias dan dengan demikian tidak bisaberpura-pura menjadi pengamat obyektif dengan penelitian. arakteristik penting lain adalah

    bahwa meluasnya penggunaan metode kualitatif dalam didikan kation relatif baru dating

    sebagian besar untuk tahun 0126 dengan perkembangan yang sedang berlangsung dalam

    metodologi dan pedoman pelaporan &$en;in 566:'. Relatif kebaruan metodologi ini juga

    berarti bahwa norma-norma profesional berdampak penelitian termasuk- ing standar bukti

    pendanaan masalah dan editorialpraktek yang berkembang &lihat misalnya Pipi 566>,

    *reeman deMarrais Preissle Roulston < #t.Pierre 5667'. $eskripsi baik kualitatif metode

    muncul dalam )ogdan dan )iklen &566?' +reswell &0112' $en;in dan =incoln &566>'

    Miles dan 9uberman &011@' dan Patton &5665'. 8da beberapa kategorisasi penelitian desain

    dalam penelitian kualitatif dan tidak ada yang uniersally disepakati &lihat misalnya $en;in

    < =incoln 566>'. +reswell &566?' tercatat lima strategi Permintaan dalam penelitiankualitatif yang saya memperlakukan sebagai sindrom onymous dengan desain penelitian3

    narasi phe #tudi nomenological studi grounded theory etnografi dan studi kasus. +reswell

    juga dijelaskan enam fase tertanam dalam setiap penelitian desain yang lebih spesifik

    daripada yang disarankan oleh +rotty &0112' tapi masih mencakup hampir semua aspek

    studi3 &0' filosofis atau teoritis perspektif, &5' pengantar studi yang termasuk tujuan dan

    penelitian pertanyaan, &?' datacollection, &@' analisis data &>' =aporan tertulis ing, dan &:'

    standar kualitas dan erifikasi. Aurnal yang mempublikasikan metodologi kualitatif makalah

    dan studi penelitian kualitatif di pendidikan tion termasuk Penelitian Kualitatif,

    KualitatifPermintaan, Lapangan Metode, Amerika PendidikanJurnal Penelitian, Peneliti

    Pendidikan,danInternational Journal of Studies kualitatif diPendidikan.+ontoh penggunaan

    kualitatif

    desain penelitian yang disediakan oleh !ahap dan Maple &011:' yang menggunakan desain

    narasi untuk menggambarkan pengalaman perempuan yang meraih bache- sebuah lor atau

    gelar master di bidang matematika dan memilih untuk mendapatkan gelar doktor dalam

    pendidikan, Gaines &566>' yang menjelajahi proses menafsirkan wawancara dan media yang

    dikumpulkan selama penulis kunjungan ke India dengan cara yang memperhitungkan nya

    identitas, 9arry #turges dan lingner &566>' yang menggunakan metode grounded theory

    untuk mengembangkan teori memberikan perspektif baru pada representasi etnis di

    pendidikan khusus, +oklat &5661' yang mempelajari perspektif uniersitas siswa

    diidentifikasi sebagai pembelajaran dinonaktifkan, dan +hubbuck dan Bembylas &5662' yang

    diperiksa perspektif dan pengajaran praktek emosional seorang guru pemula Putih di sekolahperkotaan. #tudi-studi ini mencerminkan beberapa-ciri penting membangun struktur dari

    penelitian kualitatif termasuk fokus pada menemukan dan memahami eksperimen yang

    ences perspektif dan pengalaman peserta melalui berbagai strategi penyelidikan. -Penelitian

    yang ies juga dilakukan dalam pengaturan naturalistik di yang in%uiry adalah fleksibel dan

    dipandu oleh par-komentar ticipants C yang dalam beberapa kasus digunakan untuk

    membangun penjelasan dari mereka pandangan dan perspektif. #ebuah fitur penting dari

    beberapa dari studi ini adalah mereka menggunakan unsur-unsur strategi yang berbeda

    penyelidikan.

    Kuantitati Metode Penelitian

    Metode penelitian kuantitatif berusaha memaksimalkan objektiitas peniruan dangeneralisasi i;ibility temuan dan biasanya tertarik ramalan. !erpisahkan untuk pendekatan ini

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    4/42

    adalah eDpec- yang tasi yang peneliti akan menyisihkan nya pengalaman persepsi dan bias

    untuk memastikan objektiitas dalam melakukan studi dan kesimpulan yang ditarik. *itur

    utama dari banyak studi kuantitatif adalah penggunaan instrumen seperti tes atau surei untuk

    mengumpulkan data dan-agama 8nce pada teori probabilitas untuk menguji statistik

    hipotesis yang sesuai dengan pertanyaan penelitian bunga. Metode kuantitatif sering

    digambarkan sebagai deduktif di alam dalam arti bahwa kesimpulan dari uji hipotesisstatistik memimpin untuk kesimpulan umum tentang karakteristik dari populasi. Metode

    kuantitatif juga fre- %uently dicirikan sebagai asumsi bahwa ada single "kebenaran" yang ada

    independen dari manusia Persepsi &=incoln < Guba 012>'. !rochim dan !anah &0125'

    didefinisikan kuantitatif desain penelitian sebagai lem yang memegang proyek penelitian

    bersama-sama. #()89 desain digunakan untuk struktur penelitian menunjukkan

    bagaimana semua bagian utama dari penelitian Proyek-sampel atau kelompok tindakan

    Halaman 4

    #%$

    Section aku aku aku.HaippoRtunities danchallenges diDesigning danconDuctingsaya

    nQuiRy

    perawatan atau program dan metode tugas-bekerja sama untuk mencoba untuk mengatasi

    pertanyaan penelitian pusat. &p. 0' $efinisi desain penelitian kuantitatif yang rumit oleh fakta

    bahwa istilah ini sering digunakan untuk mengidentifikasi desain reflect- eksperimental ing

    penataan independen dan dependen penyok ariabel yang terkait dengan pengumpulan data.

    ategorisasi yang lebih tua dari desain eksperimental cenderung menggunakan bahasa

    analisis arians di menggambarkan ini layout-misalnya sebuah single *aktor benar-antara

    subyek fiDed-efek desain atau desain split-petak faktorial dengan antara- dan dalam subjek

    efek &empthorne 01>5, Einer 01:5'. $eskripsi ini adalah typi- +ally terkait dengan desain

    eksperimental yang subyek secara acak ditugaskan untuk pengobatan kondisi desain

    meskipun kuasi-eksperimental di mana kelompok-kelompok utuh &misalnya 8frika 8merikadan siswa Putih' dibandingkan dan korelasi nasional desain di mana ada kelompok

    didefinisikan adalah ini juga mendapat perhatian &misalnya +ampbell

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    5/42

    erlinger &01:@', )alnaes dan +aputi &5660', Gay Mills dan 8irasian &566>', dan Metode

    Penelitian )ase &!rochim 566:' pengetahuan. ada juga beberapa jurnal yang

    mempublikasikan-metodologi kuantitatif makalah ology dan studi penelitian kuantitatif

    dalam pendidikan termasuk Metode psikologis,Jurnal Pendidikan dan Perilaku

    Statistik,Britis Journal of Matematika dan statisti! Psikologi, Amerika Pendidikan Jurnal

    Penelitian Ameri!an Journal of "valuasi, dan Pendidikan"valuasi dan AnalisisKe#i$akan.+ontoh desain penelitian kuantitatif di literatur pendidikan disediakan oleh

    9owell Eolf +ampbell dan Peterson &5665' yang menggunakan rancangan acak untuk

    mempelajari efektiitas oucher sekolah dalam meningkatkan prestasi, Fakub dan =efgren

    &566@' yang menggunakan &disconti- regresi kuasi-eksperimental nuity' desain untuk

    mempelajari hubungan antara menerima perbaikan pendidikan dan prestasi ment, dan Garner

    dan Raudenbush &0110' yang menggunakan desain korelasional untuk memeriksa relatiitas

    yang tionship antara karakteristik lingkungan dan prestasi. #tudi-studi ini mencerminkan

    beberapa fitur penting dari penelitian kuantitatif termasuk asumsi bahwa peneliti menetapkan

    selain mereka pengalaman persepsi dan bias dalam melakukan penelitian dan menggambar

    conclu- aksesi pengumpulan data yang tidak memperlakukan Peneliti sebagai instrumen

    pengumpulan data dan penggunaan pengujian hipotesis untuk membuat deduktif kesimpulantentang karakteristik populasi. #ingkatnya desain penelitian dalam pendidikan memiliki

    pindah dari ketergantungan hampir eksklusif pada %uantum metode titatie untuk pendekatan

    yang lebih berariasi yang termasuk metode penelitian kualitatif. #ebagai =incoln dan Guba

    &012>' menunjukkan baik kualitatif dan metode penelitian kuantitatif menekankan

    kebenaran konsistensi penerapan dan netralitas saat mengambil pendekatan prosedural yang

    berbeda untuk menjamin kualitas. $engan demikian mungkin akan membayangkan bahwa

    metode kualitatif dan kuantitatif telah

    Halaman 5

    !a" #$. Desain Penelitian Kualitati di / Kuantitati / Mixed Metode

    #%#

    dipromosikan menjadi peneliti pendidikan di semangat "Gunakan metodologi atau

    methodolo- gies yang &yang' yang paling tepat dan membantu untuk tujuan 8nda "tapi itu

    tidak cukup kasus. #ebaliknya penelitian pendidikan dalam terakhir dua dekade telah

    menjadi rumah bagi cukup onflik dipicu oleh pendukung masing-masing methodol-

    kelemahan ogy yang telah menekankan dari lain dalam hal epistemologi penting untuk

    bidang kesimpulan diperbolehkan dan sebagainya &lihat misalnya )erkenkotter 0110,

    $en;in < =incoln 566>, Gage 0121, 9owe 5661, MaDwell 566@, Morse 566:, Pegues

    5667'. )anyak perdebatan telah berguna dalam mendorong peneliti untuk berpikir tentang

    fitur metodologi ini penting untuk pekerjaan mereka dan untuk lapangan &misalnya Moss et

    al. 5661' namun kadang berkepanjangan dan kali alam sengit telah mungkin diminta banyakpeneliti untuk hanya duduk itu.

    Metode ca&puran

    ualitatif s kuantitatif memiliki perdebatan bertepatan dengan pesatnya perkembangan

    campuran metode yang menggabungkan kualitatif dan kuantitas produk metode tatie

    dengan cara yang seolah-olah menjembatani mereka perbedaan dalam pelayanan menangani

    penelitian pertanyaan. 8kar metode campuran yang typi- +ally ditelusuri ke multi-sifat

    multi-metode Pendekatan dari +ampbell dan *iske &01>1 dikutip dalam !eddlie , C+athain Murphy < /icoll 5667, 8pa Pekerjaan

    +learinghouse 5662'. Meskipun popularitas mereka tidak ada kesepakatan luas pada apa

    consti- tutes metode penelitian campuran &Morse 5606'. ntuk

    Halaman 6

    #%'Sectionaku aku aku.HaippoRtunities danchallenges diDesigning danconDuctingsaya

    nQuiRy

    Misalnya beberapa penulis bersikeras bahwa meth- campuran Penelitian ods adalah setiap

    studi dengan baik kualitatif dan data kuantitatif sedangkan penulis lain mengatakan metode

    campuran studi harus memiliki meth- campuran Pertanyaan ods baik kualitatif dan

    kuantitatif analisis dan kesimpulan yang terintegrasi &!ashakkori 5661'. 8da juga perbedaan

    pendapat tentang berbagai aspek metode campuran seperti ketika pencampuran harus terjadi

    &misalnya pada titik desain- ing studi selama pengumpulan data selama data yang analisis

    dan 4 atau pada titik interpretasi'. Masih penulis lain telah mengkritik seluruh Ide metode

    campuran &$en;in 566:, #ale =ohfeld < )ra;il 5665, #mith < 9odkinson 566>' kritikyang kadang-kadang dibingkai dalam hal respon pendukung khususnya untuk para a lar

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    7/42

    "sikap" argumen untuk mencampur metode3 sikap murni sikap pragmatis dan sikap dialektis

    &Greene < +aracelli 0117, Aohnson < nwuegbu;ie 566@, =awren;

    keprihatinan bahwa metode campuran desain meninggalkan kualitatif metode dalam posisi

    yang sekunder metode kuantitatif &$en;in 566:, Giddings 566:, Fin 566:'. Para peneliti

    yang mengambil sikap pragmatis berpendapat bahwa perbedaan paradigma independen dari

    dan karenanya dapat digunakan bersama dengan satu sama lain dalam pelayanan mengatasi-

    pertanyaan yang tion &Aohnson < nwuegbu;ie 566@, Morgan 5667'. Eheeldon &5606'

    merangkum pandangan ini3 )ukan mengandalkan penalaran deduktif dan tempat umum untuk

    mencapai kesimpulan tertentu atau pendekatan induktif yang berusaha umum kesimpulan

    berdasarkan tempat tertentu pragmatisme memungkinkan untuk abduktif lebih fleksibel

    pendekatan. $engan berfokus pada pemecahan praktis masalah perdebatan tentang

    keberadaan obyektif "kebenaran" atau nilai subjektif persepsi dapat bergunamengesampingkan. $engan demikian pragmatis tidak punya masalah dengan menegaskan

    kedua bahwa ada satu "dunia nyata" dan bahwa semua indiidu memiliki interpretasi mereka

    sendiri yang unik dunia itu. &p. 22' /amun sikap pragmatis juga memiliki crit- nya ics

    &Mertens 566?,. #ale et al 5665'. $ialektis peneliti berpendapat bahwa beberapa paradigma

    yang kompatibel dan harus digunakan tetapi mereka berbeda- ences dan implikasi untuk

    penelitian harus dibuat jelas &Greene < +aracelli 0117' .Ini adalah impor- tant untuk

    menekankan bahwa seorang peneliti yang mengadopsi sikap dialektika akan hal lain

    dianggap sama menggambar pada prosedur yang sama untuk mencampur sebagai salah satu

    mengadopsi sikap pragmatis. Masalah sikap tentu tidak diselesaikan dan pembangunan

    tambahan 8#I9 tentang topik ini terus muncul di metode campuran literatur &misalnya

    desain sikap dari +reswell < Plano +lark 5667'. Metode campuran desain tampak terutama

    tegas berakar dalam literatur ealuasi. 8wal kertas penting di daerah ini adalah Greene

    +aracelli dan Graham &0121' yang menyoroti lima besar tujuan &atau pembenaran untuk'

    suatu campuran meth- ods ealuation.ne adalah triangulasi yang-contoh ines konsistensi

    temuan seperti yang diperoleh melalui instrumen yang berbeda dan yang mungkin termasuk

    wawancara dan surei. Menurut Green et al. !riangulasi meningkatkan kemungkinan bahwa

    ancaman terhadap kesimpulan akan con- dikendalikan. !ujuan kedua adalah saling

    melengkapi yang menggunakan data kualitatif dan kuantitatif 9asil untuk menilai tumpang

    tindih tetapi aspek yang berbeda dari fenomena yang diteliti &misalnya kelas di pengamatan

    surei' dan ketiga adalah pembangunan ment di mana hasil dari satu metode influ- (nce

    metode berikutnya atau langkah-langkah dalam penelitian, misalnya wawancara dengan gurumungkin nyarankan- gest bahwa tambahan end-of-tahun menjadi penilaian menambahkan.

    #ebuah tujuan keempat dari metode campuran ealuasi inisiasi di mana hasil dari satu

    Metode tantangan hasil lain atau merangsang baru arah untuk penelitian, misalnya guru

    wawancara mungkin menantang hasil yang diberikan oleh administrator di distrik sekolah.

    elima dan !ujuan terakhir adalah ekspansi yang dapat memperjelas hasil atau menambah

    kekayaan dengan temuan. #ejumlah kerangka kerja untuk campuran meth- ods telah muncul

    dalam literatur ini banyak yang telah dibangun pada karya Greene et al. &0121' &+aracelli ' to the process of how academic counselors make deci- sions and used this model and a

    case study approach &+reswell 0112'. /orman purposiely sampled : counselors and 5@

    students and col- lected data ia in-depth interiews with the counselors and students that

    included their obserations of the adising process and high school and college records of

    studentsC mathe- matics course taking and grades. /orman reported that the counselorsgenerally misinter- preted the mathematics portions of high school transcripts which had

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    10/42

    important implications for adising a student on which college mathe- matics course to begin

    with.*or eDamplea student whose transcript indicated that his or her highest completed high

    school mathematics course was JIntegrated Mathematics IKL a standards-based course was

    generally adised to start with a precalculus mathematics course een though Integrated

    Mathematics IK is a precalculus course and the student should hae been adised to enroll in

    +alculus I. /orman also reported that counselors who looked at transcripts of students whocompleted a traditional high school mathematics curriculum in which the highest

    mathematics course completed was listed as precalculus were more likely to recom- mend

    that a student enroll in +alculus I./orman suggested that counselor iews toward standards-

    based curricula may be related to working in a mathematics department because mathematics

    departments hae generally been %uite critical of standards-based curricula &Roitman 0111,

    #choenfeld 566@'. /orman &5662' also collected and analy;ed %uantitatie data for a sample

    of more than 0666 college freshmen that included informa- tion on the high school

    mathematics curricu- lum they completed, their score on a college mathematics placement

    eDam, and the difficulty of their first college mathematics course which was captured using a

    @-point =ikert ariable &0 a course that should hae been completed in high school which

    is sometimes referred to as a deelopmental course, @ a course whose dif- ficulty eDceededthat of +alculus I'. 9asil of these analyses suggested that the curriculum a student completed

    was related to his or her mathematics placement score and the difficulty leel of the studentCs

    first college mathematics !entu saja. In particular students who completed a standards-based

    high school mathematics curric- ulum were more likely to enroll in a less difficult

    Halaman 9

    *hapter #$. Research Design in Qualitative/Quantitative/Mixed Methods

    #%%

    college mathematics course compared to students who completed a traditional curriculum.

    /ormanCs &5662' study reflects seeral eDem- plary features of miDed methods

    researchinclud- ing using a case study approach that focused on discoering andunderstanding the eDperiences perspecties and thoughts of counselors and stu- dents and

    the presence of %uantitatie data high- lighting the possible conse%uences of counselorsC

    misunderstanding the information on a studentNs high school transcript. Ehile the eDact

    timing of the %uantitatie and %ualitatie data collection in /ormanCs study was unclear it

    appears the data were collected relatiely close in time suggesting that the results of one data

    collection did not influence the other data collection. /amun main %uestion focused on

    counselors and col- lecting and analy;ing data for counselors first intentionally using this

    information to guide the collection and analysis of %uantitatie data for students and

    integrating the two sets of findings suggest that a se%uential eDploratory design would hae

    been of alue. *or eDample informa- tion obtained from the %ualitatie data about

    counselorsC misunderstanding of high school mathematics course taking could hae been usedin deeloping %uestions for a student sureysuch as on whether students thought their

    counselor misunderstood their high school course taking. 8 third miDed methods approach is

    these%uential transformative design,in which either %ualitatie or %uantitatie data may be

    collected pertama. 9ere the theoretical perspectie underly- ing the methodology is critical

    to the conduct of the study and the chosen methods should sere the theoretical perspectie.

    nce again %ualita- tie and %uantitatie data are analy;ed sepa- rately and the findings are

    integrated during the interpretation phase. !his approach is often used to ensure that the

    iews and perspecties of a dierse range of participants are represented or when a deeper

    understanding of a process that is changing as a result of being studied is sought. Its strengths

    and weaknesses are similar to those of the se%uential eDplanatory design. *or eDample the

    decision-making model used by /orman &5662' suggested that collecting data for studentswho are affected &sometimes adersely' by the recommendations proided by counselors is

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    11/42

    important in ealuating and improing the adising process. /orman commented that most

    students follow the recommendation of which mathematics course to take een if they

    disagree dengan itu. 9ere the study could focus on discoer- ing and understanding

    studentsCeDperiences with the adising process and its implications for their college

    eDperience.8 case study approachusing a purposiely chosen sample of students who began

    their college mathematics course taking at different difficulty leels would be appropriate.Information obtained from interiews and stu- dent academic records could be used to inform

    the construction of a surey to be sent to a repre- sentatie sample of students. !he surey

    results could be used to improe decision making by counselors and to enhance

    generali;ability. 8 fourth miDed methods approach is the !on&!urrent triangulation design,

    which is used when the focus is on confirming cross-alidating or corroborating findings

    from a single study. Hualitatie and %uantitatie data are collected concurrentlysuch that

    weaknesses of one kind of data are ideally offset by strengths of the other kind. !ypically

    e%ual weight is gien to the two kinds of data in miDing the findings although one kind of

    data can be weighted more heaily. !he %ualitatie and %uantitatie data are analy;ed

    separately and miDing takes place when the find- ings are interpreted. Important strengths of

    this approach are the ability to maDimi;e the infor- mation proided by a single study foreDample when interest is in cross-alidation and a shorter data collection period compared

    to the se%uential data collection approaches. Important weak- nesses include the additional

    compleDity associ- ated with collecting %ualitatie and %uantitatie data at the same time and

    the eDpertise needed to usefully apply both methods. $iscrepancies between the %ualitatie

    and %uantitatie findings may also be difficult to reconcile. In the 9owell et al. &5665' study

    the primary finding was that the achieement of 8frican 8merican students who receied a

    oucher to attend priate school was on aerage higher than that of 8frican 8merican

    students who did not receie a oucher and that this difference did not emerge for other

    student groups.8dopting a con- current triangulation design could proide an eDplanation for

    these findings by collecting %uali- tatie data in the form of interiews with parents of

    students who did and did not receie a oucher and %uantitatie data in the form of student

    test scores and background information. Ini akan offer an opportunity to corroborate findings

    from this study with respect to the improed achieement of 8frican 8merican students but

    Halaman 10

    #%+ Section aku aku aku. HaippoRtunities anDchallenges inDesigning

    anDconDuctingsaya nQuiRy

    not other students. *or eDample a plausible out- come of concurrent data collection is that the

    %ualitatie data suggest that parents of 8frican 8merican students appeared to be more

    commit- ted to and enthusiastic about their studentsC education in general and the oucherprogram in particular than parents of other students and that this enthusiasm persisted

    throughout the school year &8chieement tests used in this study were gien at the end of the

    school year.'. $idalam instance the %ualitatie and %uantitatie infor- mation would proide

    corroborating eidence that the improed achieement of 8frican 8merican students could be

    attributed in part to receiing a oucher and enrolling in a priate school and in part to the

    support encourage- ment and motiation of studentsC parents. !he fifth approach is the

    !on!urrent nesteddesign,in which %ualitatie and %uantitatie data are collected concurrently

    and analy;ed together during the analysis phase. =ebih besar weight is gien to one kind of

    data in the sense that one kind of data is typically embedded in yang lain. 9oweer there

    may or may not be a guiding theoretical perspectie.8 popular appli- cation of this approach

    is with multileel struc- tures &!ashakkori < !eddlie 566?' in which different leels or unitsof an organi;ation are dipelajari. #trengths of this approach include the shorter data collection

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    12/42

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    13/42

    )ryk5665'.!hese analyses could eDplore the impact of nesting and the impact of counselor

    ariables on student data and proide a powerful eDplanation of how and potentially why

    students enrolled in a particular college mathe- matics course. !he siDth miDed methods

    approach is the !on&!urrent transformative design.8s with the se%uen- tial transformatie

    designthere is a clearly defined theoretical perspectie that guides the methodol- ogy. In this

    approach %ualitatie and %uantitatie data are collected concurrently and can be weightede%ually or une%ually during the integration of !emuan. Hualitatie and %uantitatie data are

    typically miDed during the analysis phase.#trengths include a shorter data collection period

    whereas weaknesses include the need to transform data so that it can be miDed in the analysis

    phase and dif- ficulties in reconciling conflicting results using %ualitatie and %uantitatie

    data. 8s an eDample of applying a concurrent transformatie design /orman &5662' reported

    eidence that some students were unhappy with the recommendations they receied but felt

    pow- erless to change the process. #uppose that the goal of the study was to deelop a theory

    eDplain- ing this sense of powerlessness in ways that could improe the adising process.

    /ormanCs adapta- tion of the Miller and Miller &566>' decision- making model includes a

    component that calls for students who are concerned about their rec- ommendation to be

    heard. !his could take the form of collecting %ualitatie information from students reflectingtheir eDperiences with the adising process and their iew of its impact on their mathematics

    course taking and simultane- ously collecting %ualitatie data for a group of konselor.

    $eeloping a theory eDplaining the studentCs feelings of powerlessness would re%uire that

    student and counselor responses be carefully eDamined and categori;ed to suggest themes to

    guide and inform the construction of the theory. !his information could then be used to

    inform the construction of a surey that would proide %uantitatie information for a

    representatie sam- ple of newly admitted college students to enhance generalisasi. !he

    strengths and weaknesses of this approach are similar to those of the other concurrent

    approaches. ther eDamples of miDed methods studies include )uck +ook Huigley

    (astwood and =ucas &5661', $ay #ammons and Gu &5662', nwuegbu;ie )ustamante and

    /elson &5606', and nwuegbu;ie et al. &5667'. Good descrip- tions of miDed methods can be

    found in )a;eley &5661' +reswell &566?' +reswell and Plano +lark &5667' Greene &5667'

    Reichardt and Rallis &011@' and !ashakkori and !eddlie &566?'. !he relatie newness of

    miDed methods means that journals that concentrate on pub- lishing miDed methods

    methodology papers and miDed methods studies in education are berkembang. !hey currently

    include theJournal ofMi'ed Metods (esear!, International Journalof Multiple (esear!

    Approa!es, )ualitative (esear! Journal,Ameri!an "du!ational

    (esear!Journal,"du!ational (esear!er,and"du!ational "valuation and Poli!y Analysis.In

    sum the miDed methods approach offers a collection of fleDible research designs that seem

    well suited to support rigorous eDaminations of promising ideas. !he siD designs of +reswell

    &566?' draw on the strengths of %ualitatie and %uantitatie methods to enhance in%uiry inways unlikely to occur with singular applications of these methods. #till it is important to

    emphasi;e that miDed methods design continues to face a number of significant challenges

    &)ryman 5667, +reswell 5661, =awren; < 9uffman 5665, !ashakkori 5661'. ne

    important challenge is resoling out- standing disagreements oer appropriate para- digms.

    !he miDed methods literature contains a number of Jmission accomplishedLstatements

    implying that important philosophical differ- ences hae been resoled primarily by adopting

    a pragmatic stance &eg +arey 011?, +reswell < Plano +lark 5667, 9aase < Meyers 0122,

    !ashakkori < !eddlie 566?'. Fet it is clear that important differences remain andelowski

    5660, Fin 566:'. 8 related challenge is achie- ing better agreement on what characteri;es a

    miDed methods study and its components. ntuk eDample *atare the components of a

    miDed methods study &research %uestions design

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    14/42

    Halaman 12

    #%-

    data collection data analyses interpretation' *enshould the components be miDed and

    o*&and how much' should they be miDed to justify the label miDed methods &see eg

    )a;eley 5661, )ryman 5667, Aohnson < nwuegbu;ie 566@, Morse 5606, !ashakkori

    gular use of %ualitatie or %uantitatie methods in a study in faor of a miDed methods

    approach. Part of the challenge of embracing more fleD- ibility in research designs is

    technical to the eDtent that it re%uires additional eDpertise on the part of a researcher orsuggests the need to deelop new research designs. #ebagai contoh researcher may plan to

    conduct a %uantitatie study to compare learning outcomes of a treat- ment group whose

    members receie a promising interention against those of a control group using longitudinal

    data. 9oweer in the serice of deeloping a better understanding of the interention the

    researcher may decide to add a preliminary component to the study in which single-subject

    methods &ratochwill < =ein 0115' will be used to eDamine the learning tra- jectories of a

    small number of purposiely sam- pled participants. !his would re%uire eDpertise in single-

    subject methodology that a researcher may or may not possess. #imilar eDamples can be

    constructed for %ualitatie and miDed meth- ods studies. #till the greater challenge for many

    research- ers is likely to be modifying personal norms defining scholarship and the ade%uacy

    of a contribution to the field. !his may re%uire a researcher to step outside the methodologicalboundaries he or she has been trained to honor and in some cases enforce and to embrace

    research designs the researcher may hae been taught and hae taught others are inferior.

    (mbracing other methodologies in ways that cross disciplinary funding and publication

    lines for eDample sering as a member of an multidisciplinary research team that includes

    &and alues' indiiduals with eDpertise in %uali- tatie or %uantitatie methods will re%uire a

    certain amount of risk taking but will help to moe the field toward more fleDible designs.

    Se!ond,researchers can work &or continue to work' to modify professional norms in their

    roles as authors manuscript reiewers journal editors panel members ealuating grant pro-

    posals and so forth to allow a greater range of studies and findings to be supported. Ini akan

    re%uire an artful balancing between encourag- ing risk taking &eg studies employing innoa-

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    15/42

    tie interentions or emerging methods of %ualitatie in%uiry' and the need to satisfy

    standards of research design and reporting that help to ensure the integrity of study results.

    Interestingly there is growing eidence of sup- port for doing just this. #ome of this eidence

    appears in studies published in journals such as theAmeri!an "du!ational (esear! Journal,

    "du!ational "valuation and Poli!y Analysis,and theJournal of Mi'ed Metods (esear!,

    that rely on multidisciplinary research teams. ther ei- dence is in the form of new fundingprograms at the # $epartment of (ducation &primarily

    Halaman 13

    *hapter #$. Research Design in Qualitative/Quantitative/Mixed Methods

    #%

    supporting %uantitatie research' such as the Race to the !op and Inesting in Innoation

    which focus on innoatie and ambitious approaches to educational change that draw on

    multiple disciplines. 8n especially important challenge for modi- fying professional norms

    lies in the practices of educational research journals that typically pub- lish %uantitatie

    studies. !he filters enforced by the editorial process of these journals often rein- force

    eDisting and narrow professional norms. (idence of their success can be found in many

    %uantitatie meta-analyses. In a typical %uantita- tie meta-analysis a sample of published

    and unpublished studies of a common phenomenon are collected eDamined and combinedO

    for eDample studies that hae introduced interen- tions &ouchers increased teacher

    professional deelopment' designed to reduce the achiee- ment gap between 8frican

    8merican and Ehite students.Kariables capturing key features of each study such as sample

    si;e percentage of the sample that was 8frican 8merican and charac- teristics of the

    interentions are deeloped and each study is coded using these ariables &+ooper 9edges

    < Kalentine 5661'. Huantitatie analy- ses of the resulting data typically proide ei- dence

    of the impact of editorial practices. *or eDamplelimiting the analyses to published studies

    will typically reeal that all &or irtually all' reported at least one statistically significantresult &see eg )orman 9ewes erman < )rown 566?, *ukkink < de Glopper 0112,

    Purdie 9attie < +arroll 5665' a reflection of the well- documented practice of publishing

    studies with statistically significant findings &)o;arth < Roberts 0175, 9ewitt Mitchell 1' !hese are important and intriguing %uestions and it seems clear that fleDible

    research designs that aggressiely support the study of promising ideas are needed to help

    answer them. +ampur aduk methods designs seem especially well suited for this task.

    Ehether miDed methods will someday enjoy an e%ual partnership with %ualitatie and

    %uantitatie research in education or supplant these methodologies is unclear and in many

    ways unimportant.Ehat is important is eDpanding the catalog of research designs aailable to

    support rigorous in%uiry. #till change that leads to more researchers embracing more fleDible

    research designs more conspicuously depends on modify- ing norms in ways that broaden therange of stud- ies and findings supported in education. )ahwa change is underway is

    undeniable, what is less clear is the scope and pace of the change. In sum miDed methods

    research offers an especially promising path toward using research design in ways that

    support rigorous eDaminations of promising educational ideas. !he time to fully embrace

    miDed methods designs has come.

    ReeRences

    8lise M8 < !eddlie +. &5606'. 8 continuation of the paradigm wars !ingkat prealensi

    methodological approaches across the social4 behaioral sciences. Journal of Mi'ed

    Metods(esear!, -,06?05:. )alnaes M. < +aputi P. &5660'.Introdu!tion to%uantitative

    resear! metods An investigative approa!.!housand aks +83 #age. )a;eley P. &5661'.MiDed methods data analysis. $i (. 9alcomb < #. 8ndrew &(ds.' Mi'ed metods resear!

    for nursing and te ealt s!ien!es &pp. 2@002'. =ondon3 Eiley-)lackwell. )erkenkotter +.

    &0110'. Paradigm debates turf wars and the conduct of sociocognitie in%uiry in

    composition. /ollege /omposition and /ommuni!ation, -0, 0>00:1. )ogdan R+ '. !he practice and politics of funded penelitian kualitatif. In / $en;in

    < F# =incoln &(ds.' !he SA7" and#ook of %ualitative resear!&?rd ed. pp. ?27@61'.

    +hubbuck #M < Bembylas M. &5662'. Itu emotional ambialence of socially just teaching3

    8 case study of a noice urban schoolteacher. Ameri!an "du!ational (esear! Journal, -4,57@?02. +ochran EG < +oD GM &01>6'. "'perimental designs. /ew Fork3 Eiley.

    +ooper 9. 9edges =K < Kalentine A+ &5661'. e and#ook of resear! syntesis and

    Halaman 15

    *hapter #$. Research Design in Qualitative/Quantitative/Mixed Methods

    #+#

    meta&analysis &5nd ed.'. /ew Fork3 Russell #age $asar. +reswell AE &0112'. )ualitative

    in%uiry and resear! design /oosing among five traditions. !housand aks +83 #age.

    +reswell AE &566?'. )ualitative, %uantitative, and mi'ed metods approa!es &5nd ed.'.

    Ribu aks +83 #age. +reswell AE &5661'. Mapping the field of miDed methods research.

    Journal of Mi'ed Metods (esear!, 2,1>062. +reswell AE < Plano +lark K= &5667'.8esigning and !ondu!ting mi'ed metods resear!. !housand aks +83 #age. +rotty M.

    &0112'. e foundations of so!ial resear! Meaning and perspe!tive in te resear! pro!ess.

    =ondon3 #age. $ay +. #ammons P. < Gu H. &5662'. Menggabungkan %ualitatie and

    %uantitatie methodologies in research on teachersC lies work and effectieness3 *rom

    integration to synergy. "du!ational (esear!er, 21, ??6?@5. $en;in / &566:'. !he

    elephant in the liing room3 r eDtending the conersation about the politics of eidence.

    )ualitative (esear!, 9,0?10:6. $en;in / < =incoln F# &566>'. Pengantar. In /

    $en;in < F# =incoln &(ds.' e SA7" and#ook of %ualitative resear!&?rd ed. pp. 051'.

    !housand aks +83 #age. *reeman M. de Marrais . Preissle A. Roulston . '. Interpreting India identity and media from the field3 (Dploring

    the communicatie nature of the eDotic other. )ualitativeIn%uiry, 33,>02>?@. Garner +.

    < Raudenbush #A &0110'. =ingkungan effects on educational attainment3 8 multileel

    analysis of the influence of pupil ability family school and neighborhood. So!iology of

    "du!ation, :-,5>05:5. Gay =R Mills G. < 8irasian PE &566>'. "du!ational resear!

    /ompeten!ies for analysis and appli!ations &2th ed.'. pper #addle Rier /A3 Merrill.

    Giddings =# &566:'. MiDed-methods research3 Positiism dressed in drag Journal of

    (esear! in 5ursing, 33&?' 01>56?. Greene A+ &5667'.Mi'ed metods in so!ial in%uiry./ew Fork3 Eiley. Greene A+ < +aracelli KA &0117'. Advan!es in mi'ed&metod

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    18/42

    evaluation e !allenges and #enefits of integrating diverse paradigms.Baru dire!tions for

    evaluation&/o. 7@ pp. 01?5'. #an *rancisco3 Aossey-)ass. Greene A+ +aracelli KA . 9iatt A* &012:'.

    #pirituality medicine and healing. Soutern Medi!al Journal, 19, 7?:7@?. 9owe R

    &5661'. Positiist dogmas rhetoric and the education science %uestion. "du!ational

    (esear!er, 2;,@52@@6. 9owell EG Eolf PA +ampbell $( < Peterson P( &5665'.

    #chool ouchers and academic performance3 Results from three randomi;ed field trials.

    Journal of Poli!y Analysis and Management, 03,010507. Aacob )8 < =efgren =. &566@'.

    Remedial education and student achieement3 8 regression discontinuity analysis.(evie* of

    "!onomi!s and Statisti!s, ;:,55:5@@. Aohnson R) < nwuegbu;ie 8A &566@'. +ampuraduk methods research3 8 research paradigm whose time has come. "du!ational (esear!er,

    22,0@5:. Aohnson R) < !urner =8 &566?'. $ata collection strategies in miDed methods

    research. In 8. !ashakkori < +. !eddlie &(ds.'6and#ook of mi'ed metods in so!ial and

    #eavioral resear! &pp. 517?01'. !housand aks +83 #age. empthorne . &01>5'.

    8esign and analysis of e'periments./ew Fork3 Eiley. erlinger *. &01:@'. +oundations of

    #eavioral resear!. /ew Fork3 9olt. ratochwill !R < =ein AR &0115'. Single&!ase

    resear! design and analysis 5e* dire!tions for psy!ology and edu!ation.9illsdale /A3

    =awrence (rlbaum.

    Halaman 16

    #+'

    =awren; *. < 9uffman $. &5665'. !he archipelago approach to miDed method ealuation.

    Ameri!an Journal of "valuation, 02,??0??2. =eech /= Morgan G8 Eang A. < Gliner

    A8 &5606'. #tatistical significance and eidence- based approach3 Implications of non-

    significant results.Journal of 5on- signifi!ant (esults in "du!ation, 3,005. =eech /= @6@. Morse AM &5606'. #imultan dan sekuensial %ualitatie miDed

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    19/42

    method designs. )ualitative In%uiry, 3:,@2?@10. Moss P8 Phillips $+ (rickson *$

    *loden R( =ather P8 < #chneider )= &5661'. =earning from our differences3 8 dialogue

    across perspecties on %uality in education research."du!ational (esear!er, 2;,>60>07.

    $ewan Riset /asional. &5665'. S!ientifi! resear! in edu!ation&RA #haelson < =. !owne

    (ds.'. +ommittee on #cientific Principles for (ducation Research. Eashington $+3 /ational

    8kademi Press. $ewan Riset /asional. &566>'.Advan!ing s!ientifi! resear! in edu!ation&=.!owne == Eise < !M Einters (ds.'. omite on Research in (ducation. Eashington $+3

    8kademi Pers /asional. /ewman I. Ridenour +# /ewman +. < $eMarco GMP &566?'.

    8 typology of research purposes and its relationship to miDed metode. In 8. !ashakkori < +.

    !eddlie &(ds.'6and#ook of mi'ed metods in so!ial and #eavioral resear!&pp. 0:7022'.

    Ribu aks +83 #age. /orman . &5662'.6ig s!ool matemati!s

    !urri!ulum and te pro!ess and a!!ura!y of initial matemati!s pla!ement for students *o

    are admitted into one of te S"M programs at a resear! institution. npublished

    dissertation niersity of Minnesota !win +ities. C+athain 8. &5661'. MiDed methods

    research in the health sciences3 8 %uiet reolution.Journal of Mi'ed Metods (esear!, 2,?

    :. C+athain 8. Murphy (. < /icoll A. &5667'. Mengapa and how miDed methods

    research is undertaken in health serices research in (ngland3 8 miDed methods study. BM/6ealt Servi!es (esear!, 1. 8ailable at http344www.biomedcentral .com40@75-:1:?4742>

    nwuegbu;ie 8A )ustamante RM < /elson A8 &5606'. MiDed research as a tool for

    deeloping %uantitatie instruments. Journal of Mi'ed Metods (esear!, -, >:72.

    nwuegbu;ie 8A Eitcher 8( +ollins M! *iler A$ Eiedmaier +$ < Moore +E

    &5667'. #tudentsC perceptions of characteristics of effectie college teachers3 8 alidity study

    of a teaching ealuation form using a miDed- methods analysis. Ameri!an "du!ational

    (esear! Journal, --, 00?0:6. Patton MH &5665'. )ualitative resear! and evaluation

    metods &?rd ed.'. !housand aks +83 #age. Pedha;ur (A < #chmelkin =P &0110'.

    Measurement, design, and analysis An integrated approa!. 9illsdale /A3 =awrence

    (rlbaum. Pegues 9. &5667'. f paradigm wars3 +onstructiism objectiism and postmodern

    stratagem. "du!ational +orum, 13, ?0:??6. Plano +lark K= &5606'. !he adoption and

    practice of miDed methods3 # trends in federally funded health-related research. )ualitative

    In%uiry, 3:,@52@@6. Purdie /. 9attie A. < +arroll 8. &5665'. lasan 8 of the research on

    interentions for attention deficit hyperactiity disorder3 Ehat works

    Halaman 17

    *hapter #$. Research Design in Qualitative/Quantitative/Mixed Methods

    #+(

    best (evie* of "du!ational (esear!, 10, :011. Raudenbush #E < )ryk 8# &5665'.

    6ierar!i!al linear models Appli!ations and data analysis metods&5nd ed'. /ewbury Park

    +83 #age. Reichardt +# < Rallis #* &(ds.'. &011@'. Itu %ualitatie%uantitatie debate3/ew perspecties. 5e* 8ire!tions for Program "valuation&no. :0'. #an *rancisco3 Aossey-

    )ass. Roitman A. &0111'. )eyond the math wars. /ontemporary Issues in Matemati!s

    "du!ation, 2:,05?0?@. Rosenthal R. &0171'. !he Jfile drawer problemL and tolerance for

    null results. Psy!ologi!al Bulletin, ;:,:?2:@0. #ale A(M =ohfeld =9 < )ra;il .

    &5665'. Reisiting the %uantitatie%ualitatie debate3 Implications for miDed methods

    research. )uality > )uantity, 2:, @?>?. #andelowski M. &5660'. Real %ualitatie

    researchers do not count3 !he use of numbers in %ualitatie research.(esear! in 5ursing >

    6ealt, 0-,5?65@6. #andelowski M.Koils +I < nafl G. &5661'. Pada %uantiti;ing.

    Journal of Mi'ed Metods (esear!, 2,562555. #chneider ). +arnoy M. ilpatrick A.

    #chmidt E9 < #haelson RA &5667'. "stimating !ausal effe!ts using e'perimental and

    o#servational designs.Eashington $+3 8merican (ducational 8sosiasi Riset. #choenfeld89 &566@'. !he math wars."du!ational Poli!y, 3;,5>?52:. #hadish ER +ook !$

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    20/42

    +ampbell $! &5665'."'perimental and %uasi&e'perimental designs for generali?ed !ausal

    inferen!e &5nd ed'. )oston3 9oughton Mifflin. #mith A < 9odkinson P. &566>'.

    Relatiism criteria and politics. In / $en;in < F# =incoln &(ds.' e SA7" and#ook of

    %ualitative resear!&?rd ed. pp. 10>1?5'. !housand aks +83 #age. #tage * < Maple

    #8 &011:'. ompatibel goals3 /arraties of graduate women in the mathematics pipeline.

    Ameri!an "du!ational (esear! Journal, 22,5?>0. !ashakkori 8. &5661'. 8pakah kitasudah sampai !he state of the miDed methods community. Journal of Mi'ed Metods

    (esear!, 2,527510. !ashakkori 8. < !eddlie +. &566?'.Mi'ed metodology /om#ining

    %ualitative and %uantitative approa!es. !housand aks +83 #age. !aylor #A >' writes that the purpose of science is theory and that by

    eDtension the purpose of research is to contrib ute theories or refinements of eDisting theories

    to science. !heory is a kind of abstraction a simplification of reality that applies in similar

    circumstances and not just to the specific case at tangan. *or researchers theories focus

    attention limit choices and proide eDplanationsO characteristics that gie good theories a

    central role in research design. *or actors in the educa tional enironment they are practical

    for the same reasons. !heories about social behaior hae inherent batas. !hese are identifiedby !horngate &017:' and elaborated on by Eeick &0171'. !horngate deeloped a postulate of

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    22/42

    commensurate com pleDity in which there are tradeoffs among a theory being general, a

    theory being a!!urate,and a theory beingsimple.8 theory cannot be all three simultaneously,

    general accurate theories are not simple accurate simple theories are not general and simple

    general theories are not tepat. Eeick proides eDamples of each. $i deeloping a research

    design the theory used or the one the researcher is trying to deelop has more important

    effects on research design than does anything eDcept the topic chosen for the penelitian.!heory dries hypotheses. !he choice of a theory to use or deelop reflects the researcherCs

    interest in being general simple or accurate and shapes the study accordingly. $ari my

    obserations I would suggest that educa tional research errs on the side of being simple and

    with luck accurate. !heory is emphasi;ed in research because it proides eDplanation.

    Eithout meaningful descriptions of the situationOthat is without identifying new ideas to be

    understood and related to each other by theoriesOresearch would not moe forward.

    $esigning research that identifies the ways in which people in a gien situation iew their

    worlds is a sensible starting place for meaningful research. !anpa important things to be

    studied no theories would need to be deeloped and a rigorous methodology to estimate

    relationships based on theory would not be necessary.

    Topics and 3deas

    9ow does one go about selecting substantie issues connected by a theory !eDts coering

    research in education or the behaioral sciences

    Halaman 21

    *hapter ##. 3ntellect0 1ight0 and Shado2 in Research Design

    #+,

    hae tended to either be mute on the %uestion or proide only minimal adice &Gall )org

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    23/42

    educational research while the others hae receied little study or hae been the proince of

    speciali;ed fields &eg music'. !he topics we choose reflect a worldiew of educational

    research3 study the specific rather than the general emphasi;e topics related to linguistic and

    logical cognitie deelopment and use methods that faor such studies. ami also faor

    topics that reflect social justice cog nitie deelopment that leads students into economically

    productie lies and those topics that fit within predetermined disciplinary spe cialties. Ifeducation deals with the whole stu dent as )owen &0177' suggests it would behooe

    researchers to look for topics that consider more kinds of intelligence than those associated

    with linguistic and mathematical pemikiran. 9aing become obsessed with cer tainty in our

    research we hae eschewed broad research topics that re%uire integrating arious

    components of human intelligence in faor of being able to say with great certainty some

    thing of diminutie importance. 9ighly spe ciali;ed methodologies re%uire highly refined

    topics which proide little opportunity for transformational outcomes and the deelop ment

    of big ideas.

    *hoosing Topics

    Identifying a resear! pro#lemis the starting place for research. Research problems inoleunresoled realworld conditions or situations and it is the research problem that is nested

    between the topic and the purpose. !heoretical research problems deal with Jwe donCt know

    whyL descriptie research problems deal with Jwe donCt know whatL and practical research

    problems deal with Jwe donCt know how.L !he best researchers are attracted to uncertainty

    paradoDesanomaliescontradictionsand ambi guities in the field. !he significance of a prob

    lem is often based on the way in which it inter sects with theoretical uncertainty and practical

    pentingnya. Probably the best way of finding a problem to study is to do eDtensie reading in

    an important topical area and find out what is poorly under berdiri. Many articles contain

    sections that gie suggestions for future research and many hae

    Halaman 22

    #+-

    Section aku aku aku. HaippoRtunities anDchallenges inDesigning anDconDuctingsaya

    nQuiRy

    glaring shortcomings that suggest a problem should be reeDamined from a different perspec

    tie. #ome researchers choose a methodology and then try to find a problem to match it. Ini

    approach creates an unnecessary constraint on what is to be studied and the foolishness of

    this se%uence cannot be emphasi;ed enough. Itu cart does not pull the horse. ther

    researchers are told what to study by a superior such as an adisor a proider of resources

    an admired scholar in the field or a coinestigator. #o long as the relationship is noteDploitatie joining an ongoing research agenda has the clear adantage of the person being

    clear about what he or she will study. It has the disad antage of not learning how to define

    oneCs own masalah penelitian. !ypically a study inoles the following iteratie process3

    8pproach a topic of interest read in the area write a brief statement of purpose discuss this

    statement with others reflect read write talk and so on until a com pelling problem with a

    realistic scope has been identified and clearly eDpressed. 8 hot topic is one where there is a

    lot of interest and a great likelihood of getting a project funded finding others to participate

    in the study and publishing the results. )aik topics allow young scholars to demonstrate their

    research skills and increase the likeli hood of getting further support for their research and

    publications in their fields. ntuk better or worse it usually means becoming more

    speciali;ed. Eith luck a good topic is one the researcher loes. 9ere loveis a pas sionateattachment to the research and an enjoyment of the research process. !his attach ment should

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    24/42

    not be confused with a lack of objectiity but it inoles caring about an increased

    understanding of a topic and a will ingness to put forward the effort that results in influential

    research. 8fter deciding what to study and why such a study is worthwhile the final part of

    designing research is to decide on the processes by which the research is to be accomplished.

    Many people in thinking about research design think that they need only be concerned with

    research methodology &erlinger 017? cited in $aniel 011:'. !he following three %uestionsalways affect research design3 &0' Ehat will the researcher study &5' Ehy is the research

    penting &?' 9ow will the researcher carry out the research nly the third %uestion is related

    to methodology.

    Dierent 6pproaches to Research

    !here are many ways in which to study the same topic and these produce different results.

    Mitroff and ilmann &0172' describe four approaches to research based on a Aungian analysis

    of our predispositions to approach deci sion making and obtain information similar to the

    Myers)riggs !ype Indicator tests &Myers )riggs 01:5'. !he four approaches include the

    following3 the scientist the conceptual theorist the conceptual humanist and the indiidual

    humanist. Gien the topic !ollege student reten&tion, consider the way in which eachmethodol ogy is chosen based on the researcherCs interests is different from the other

    methods and pro duces different results. $escriptie phrases below are taken from tables in

    Mitroff and ilmann on the pages indicated. *or the s!ientist, the approach should be

    objectie causal cumulatie and progressie emphasi;ing reliability and eDternal alidity

    and separating the scientist from the obsered. #aya t aims at precise unambiguous empirical

    knowl edge using strict logic &Mitroff < ilmann0172 p. ?@'. !he norms of this approach

    are known as the +$#3 * ommunism indicating that scientific knowledge is common

    property, 5niersalism indicating that scientific knowledge should be independent of the

    personality of the indiidual scientist, D isinterestedness such that the scientist should

    obsere what happens and not adocate a theory or eDperimental outcome, dan 7rgani;ed S

    kepticism where scientists should be critical of their own and othersC ideas. &Merton

    01@54017? p. 5:1' 8n eDample of the scientific study of reten tion would be an

    organi;ational eDperiment based on the hypothesis that higherachieing students are more

    likely to remain enrolled in perguruan tinggi. #tudents would be randomly assigned to a

    treatment group or a control group. $alam treatment group students would participate in a

    Halaman 23

    *hapter ##. 3ntellect0 1ight0 and Shado2 in Research Design

    #+

    retention program such as deeloping study skills but otherwise would hae eDperiences no

    different from those of the control group. #etelah a gien time period the researcher wouldfind out whether the retention rate for students who participated in the retention program was

    sig nificantly different from that for students who tidak. !his information would be used to

    sup port or negate the hypothesis. !he !on!eptual teoristis inoled with research that is

    impersonal aluefree disinter ested imaginatie and problematic inoling multiple

    causation purposeful ambiguity and ketidakpastian. !he theorist is interested in the conflict

    between antithetical imaginatie theo ries comprehensie holistic theories and eer

    eDpanding research programs to produce con flicting schemas using dialectical and indetermi

    nate logics &Mitroff < ilmann 0172 p. >:'. #()89 theorist conducting retention

    research would proide at least two theoretical eDplanations of retention behaioruse surey

    methods to gather information analy;e the data using statistics find out whether the data

    supported one theory more than the otherand use that information to make more theories.Much of the empirical research reported in educational journals is a combination of the

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    25/42

    scientist and theoristO theory guiding social science in%uiry into educational structures and

    processes. !he !on!eptual umanist&although I findso!ial umanistto be a more accurate

    descrip tion' approaches research as a alueconstituted interested actiity that is holistic

    political and imaginatie, where multiple causations are pres ent in an uncertain and

    problematic social eni ronment, and with a deep concern for humanity. !his approach

    recogni;es the importance of the relationship between the in%uirer and the subject and has theaim of promoting human deelopment on the widest possible scale. Itu normatie outcomes

    of such research would be economic plenty aesthetic beauty and human kesejahteraan.

    #imilar to an action researcher the social humanist prefers smallgroup dynamics where both

    the in%uirer and the participants learn to know themseles better and work together to

    improe the situation &Mitroff

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    26/42

    dapat diterapkan. !o do good research the meth odology used should be appropriate for the

    problem addressed. !his is a necessary condi tion but not a sufficient one. 8n elegantly ana

    ly;ed data set that was composed of ambiguously measured data that addressed a %uestion of

    tri ial importance is not likely to enter the annals of great research. (ducational research is

    part of the social sci ence research tradition a tradition that was influenced by research in the

    natural sciences. !he natural sciences use the scientific method to sole research problems orsupport a perspec tie. !he method contains a series of se%uential steps similar to the

    following3 Identify a prob lem gather information from the literature about this %uestion

    deelop a hypothesis in the conteDt of a theory collect data related to the hypothesis analy;e

    the data and draw a conclu sion related to the truthfulness of the hypothesis and correctness

    of the theory. #cientists as logical purists build argu ments onfalsifia#ilityand the la* of te

    e'!luded middle.!his law states thatAand not&Acannot eDist simultaneously. )ut ifA stands

    for Jthis program helps students to learnL and not&Astands for Jthis program does nothelp

    students to learnL then both can be true as in the case of aptitudetreatment inter tindakan,

    that is a treatment could be effectie for a highaptitude student but not effectie for a

    lowaptitude student. If both are true then the law of the eDcluded middle is iolated and

    falsifiability cannot be demonstrated. !his sit uation is problematic for scientific researchdalam pendidikan. (ducation is not a scientifically based process partly because the term

    edu!ationis ideological and idiosyncratic much different from the term temperature.8t best

    scientific research can shed light on narrowly defined educational behaiors and researchers

    can hope forObut cannot guaranteeOa cumulatie effect. Ehen a goern ment policy

    assumes that education is e%uialent to improing the score on a test the society will not

    hae a moral compass and will not be educated. *eyerabend &011?' holds the iew that if we

    do not separate scientific research and the state as we hae separated the church and the

    state irreparable harm will be done. In the same way that social science research has imitated

    the natural sciences educational research has imitated social science research. Research in

    these areas may be separated more by the topic studied than by the rigor of the metodologi.

    8s with social science research in general educational research might pre tend a leel of

    control so as to carry out an percobaan. Researchers gie themseles sol ace that Jother

    things are e%ualL or Jother effects are randomL or Jspuriousness is not a problemL and

    proceed as if the social world were simple and understandable in the same way that the

    traditional world of the pure sci ences can be.

    6 Traditional 6pproach to Designing

    4ducational Research

    Graduate programs in education typically are arranged in departments that reflect academic

    subspecialties such as history of education soci ology of education anthropology of

    education counseling psychology eDperimental psychol ogy higher education schooladministration and curriculum and instruction. #ebagian besar programs re%uire a course in

    research design appropriate for their field. My discussion reoles around %uantitatie and

    %ualitatie approaches to research terms that reify catego ries that are themseles

    oerlapping and arbi trary but are commonly used to describe research courses. 8 simple way

    of looking at a proposal is to see how it answers the three %uestions posed earlier3 Ehat is

    this study about Ehy is the study penting 9ow will the researcher conduct the belajar !he

    study itself must coer these three %uestions and answer two additional %uestions3 Ehat did

    the researcher find Ehat do the find ings mean Research designs reole around a limited

    number of elements. !heir eDact use and eDposition ary depending on the particular

    approach taken. !he purpose and promise of these elements hae been identified and dis

    cussed by a number of teDtbooks such as those of Gall and colleagues &5665' and +reswell&5667'. !hese teDts identify many of the issues facing researchers especially those who are

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    27/42

    new to the process. 8lthough not suitable for all studies well designed %uantitatie research

    usually ad dresses the areas presented in the following garis besar3

    Halaman 25

    *hapter ##. 3ntellect0 1ight0 and Shado2 in Research Design

    #,#I. Introduction to the topic of the research #ebuah. )ackground and conteDt in which that

    topic has been studied b. Importance of studying the topic termasuk c. Practical alue of the

    study d. Research problem to be addressed e. !ujuan dari penelitian ini f. bjecties or

    %uestions to be addressed g. $efinitions and related constructs h. 8ssumptions used in the

    study saya. eterbatasan penelitian j. =ingkup penelitian II. sing the literature to build

    conceptual argumen #ebuah. Releant theories to guide the study b. !he findings of other

    researchers that identify important ariables related to the purpose c. Identification of the

    dependent ariable independent ariables and theories linking these ariables 8 8

    8. Methodology to be used in the study #ebuah. #ite sample or selection procedure for

    respondents including b. 9ow the data will be gathered c. 9ow the data will be measured d.

    9ow the data will be analy;ed e. Ehy these methods are appropriate. Q!his information

    usually completes the design of a research proposal and appears as the first part of a finished

    belajar. *inished studies also include the following. IK. !emuan #ebuah. 8 description of the

    sample actually analy;ed in the study b. $escription of the data c. !reatment of missing cases

    d. Possible or known biases e. $escription of how the researcher met the assumptions

    re%uired to use the chosen statistics f. Presentation of the data g. #upport or lack of support

    for the hypotheses or theories used h. $iscussion of the findings saya. esimpulan K.

    #ummary of the study #ebuah. Practical implications of the study b. 8reas for future research

    Hualitatie research can inole using the fie research traditions identified by +reswell

    &0112'Obiography ethnography grounded the ory case study and phenomenologyOwhich

    can be used singly or in combination. General head ings appropriate for %ualitatie studiesinclude the topic focus and purpose significance related liter

    aturemethodologypresentation of the datainter pretation of the data and conclusions.

    !erperinci headings would be similar to the following3 I. !opic to be studied #ebuah. !he

    oerall interest focusing on what will be eDplained or described b. rgani;ing metaphor &like

    grounded theory'

    c. !he mystery and the detectie

    d. 9ermeneutic elements

    e. !he significance of the study

    f. Ehy the reader should be interested.

    II. Getting information away from the source

    #ebuah. Releant literatureb. !eori

    c. *indings for content area themes foci

    or analogous situations

    8 8 8. !he selected %ualitatie method

    #ebuah. 9ow information is obtained

    b. 9ow sense is made from it

    c. #ite selection

    d. Informant selection

    e. Pengumpulan data

    f. 8nalisis data

    g. $ata ealuation

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    28/42

    Q!he methodology can be in a separate chapter part of the first chapter in an appendiD or

    woen into the chapters that present respondent information. !his section is followed by one

    or more chapters which does the following.

    IK. Present the teDt as natural answers to natural %uestions. #ebuah. Presentation formats

    include stories tables interiews documents narraties photographs ideotapes ignettesteDts of arious kinds descriptions and routines &see #chwart;man Q011? from which some

    of these headings were taken'. b. Raw data can be presented without comment presented and

    interpreted simultaneously or presented and interpreted in a later section. K. *indings

    #ebuah. esimpulan b. Recommendations for others

    Halaman 26

    #,'

    !he structure of %ualitatie studies is more idiosyncratic than the more formal structure of

    traditional %uantitatie studies. In %ualitatie research the sample is the study and the

    reasons for selecting the sample need to be emphasi;ed. Most researchers when approaching

    a topic they care about hae tentatie hypotheses about what causes what or predispositions

    to thinking that the world operates according to certain principles that also apply in this area.

    Masyarakat bias their obserations based on their eDperi ence. 8ll of us know that we hae

    certain biases and we can try to counter those biases in our research by looking for eidence

    that directly contradicts what we eDpect. In the unconscious there is a second set of biases of

    whichby defini tion we are not aware. #ometimes peer readers can help the researcher to

    discoer what is miss ing or what is inappropriately under or oer emphasi;ed in the study.

    8fter analy;ing the data in a %uantitatie study the researcher presents the findings.

    !ypically it is rather straightforward because the data to be gathered and the analyses pro

    posed for the data were specified in the proposal untuk penelitian. *or %ualitatie researchers

    the data the findings and the method might not be berbeda. !he narratie that presentsselected %uestions and answers can represent findings based on data that came from the

    method by which %uestions were deeloped. 8s the prei ous sentence suggests it is a

    conoluted process. !he presentation might reole around respon dentsC eDperiences and

    understandings a chro nology of eents or themes supported by respondentsC statements. In

    ethnographic stud ies the descriptions of lies in conteDt can stand on their own

    &=awrence=ightfoot 011>'. !hick descriptions &Geert; 017?' might pro ide greater insight

    into the education of a stu dent in a school than would an analysis of ariabel. In most

    studies some analysis of the descriptions is eDpected. !his predisposition is part of the legacy

    of pragmatism, peneliti in education are eDpected to identify how knowledge gained from the

    study can improe professional practice. In designing a study the background con teDt

    importance of the topic and presumed practical alue of the study come from the lit eraturewritten about the topic or analogous literatures in similar fields.*or eDamplethe study of

    college student retention can be considered to be analogous to the study of turnoer in work

    organi;ations and the literature in one area can be used to reinforce the literature in the other

    area &)ean 0126'. !he use of literature in %uan titatie studies howeer can differ

    substantially from the use of literature in %ualitatie studies. In a %uantitatie study the

    literature is used to identify the importance of the dependent ariable releant independent

    ariables and theories that bind these factors together to jus tify the use of statistical

    procedures and to pro ide a conteDt for the discussion. In %ualitatie studies a premium is

    placed on the ability to see what is before the researcher. ur ability to obsere is both

    heightened and diminished by our prior knowledge and eDpectations &)ean 0117'. It is

    heightened by making ourseles aware of important details to obsere and it is diminishedbecause we focus only on those rincian. $ue to the preconceied notions of the researcher

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    29/42

    those factors actually influencing the respondentsC world might not be identified. Ehen the

    literature shapes the way in which we iew the world what is actually before us is replaced

    by what we eDpect to see. 8 reiew of the literature as a standalone section summari;ing

    research in the topical area makes little sense. !he literature as a com pendium of related

    information should be used to adance arguments related to the importance of the subject. It

    should identify topical areas that are either well or poorly understood iden tify and describereleant theories identify and describe appropriate methodologies to study the topic describe

    dependent and independent ariables if releant proide definitions and proide a conteDt to

    discuss the findings from pembelajaran. Google #cholar $issertation 8bstracts International

    (RI+ $ocuments the I#I Eeb of nowledge and the proceedings of releant professional

    organi;ations all can be used to access current research. 8 condemning retort is that the

    literature in a study is dated.!his phrase has some interesting subteDts. !he first assump tion

    is that the most recent research is the best research and that preious research is irreleant. 8

    second assumption is that all research is of limited generali;ability oer time so that if it is

    older than say > years it is irreleant. In either case dated research is of marginal alue.

    leh eDtension the research that a person is currently conducting is also of marginal alue

    because it

    Halaman 27

    *hapter ##. 3ntellect0 1ight0 and Shado2 in Research Design

    #,(

    will be useful for only > years. !his planned obsolescence of research becomes a justification

    for a fren;ied increase in the rate of publication and is counterproductie in terms of

    identifying important and durable ideas in the field. =iterature should not be weighed nor

    consid ered dated after > years. In traditional %uantitatie research the topic contains the

    dependent ariable and the factors associated with it identify the independent ari ables that

    hae been found to hae important effects on the dependent ariable. In studies that are not

    codificationsOthat is not eDtensie reiews of the literature for the heuristic pur pose of

    organi;ing what is known about a topicOciting the literature should be done for the purpose

    of building an argument not sim ply to show familiarity with the canon. #ince the 01:6s the

    number of statistical analyses aailable for researchers to include in their designs has

    increased dramatically. =ima commercial statistical packages bear the initials #8# #P##

    )M$P G=IM and 9=M. !he deel opment of these statistical packages has allowed eer

    more compleD analyses to be performed. /ational data sets from the /ational +enter for

    (ducational #tatistics &/+(#' and other sources hae proided the opportunity to bring order

    to ast amounts of data. *or the description of largescale phenom ena these data sets can be

    ery aluable. ntuk analy;ing the causes of behaior howeer the attempt to gain a broad

    ision masks indiidual or smallgroup differences. =ongitudinal studies almost always sufferfrom decay, that is mea sures may differ from year to year and respon dents drop out of the

    study. #o the comparisons from year to year might not be the result of what people report,

    rather they might be the result of changes in who is doing the reporting. !he aailability of

    data and the means to analy;e them raised the leel of eDpectation in some journals that such

    analyses should be the norm. Ehat is certain is that during the past >6 years the

    sophistication of analyses has increased. !he literature shifted from normed sureys that

    reported fre%uencies to chi s%uares to analyses of ariance &8/K8s' and simple

    correlations to factor analysis and multiple regression to causal modeling with ordinary least

    s%uares path analysis to maDi mum likelihood used in linear structural rela tions &=I#R(='

    modeling and to generali;ed linear modeling &G=IM' and hierarchical linear modeling

    &9=M'. !he increase in compleDity is associated with an increase in agitated eDchangesbetween statis ticians about whose method is correct. #ebuah improed methodology has not

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    30/42

    been matched by these studies becoming more influential in policy making or practice

    &e;ar 5666'. Itu debate is sometimes inisible to the public tak ing place between the

    author of a piece of research and the consulting editors who reiew the research and

    sometimes it appears in jour nals such as the"du!ational (esear!er.

    DataHuantitatie studies re%uire data that can be used in statistical analyses. !he sources of data

    can ary widelyOhistorical documents goern mental records organi;ational records inter

    iews standardi;ed sureys %uestionnaires deeloped as part of the research protocol for a

    particular study unobtrusie measures obser ations participant obseration and so on. Itu

    %uality of the research depends on the %uality of the data analy;ed, data analysis has only a

    sec ondary influence. !he %uality of the data aries greatly. )aik research design re%uires that

    the researcher understand the strengths and weaknesses of the data. 9istorical data can reflect

    the biases and ideological preferences of those who recorded it. People who proide data can

    intentionally dis tort it to put themseles in a better light for eDample reporting that they had

    higher grades than they actually did. #urey data might come from a biased sample reflecting

    only the eDperi ences of highsocioeconomic status respondents. Huestions in a surey mightbe ambiguously written or a single item might contain two %uestions with different answers

    for eDample J9ow satisfied are you with your salary and fringe benefitsL #urey data that

    re%uire a forcedchoice response might not represent the real interests of the respondent. 8

    respondent might hae no opinion on most of the %uestions and refuse to answer them. ther

    respondents might not want to reeal personal information and so might misrepresent their

    actual incomes whether they hae eer plagiari;ed or how much they use drugs or alcohol.

    walaupun %uestionnaire is not missing any data the data proided might be intentionally

    inaccurate.

    Halaman 28

    #,)In other cases respondents might not under stand the %uestions might not care about the

    answers gien or might become fatigued while filling out the %uestionnaire so that the

    accuracy of the responses is different for the beginning and the end of the %uestionnaire. 8

    wellwritten %uestion should reflect one bit of information about the respondent

    unambiguously and reli ably and the answer to the %uestion should match obserable facts. It

    is acceptable to use problematic data if the analyst understands and acknowledges the prob

    lems that eDist in the data.*or eDamplea data set might not be random but might be

    completely representatie of one subgroup in the population dipelajari. !he bias in this

    sample can make con clusions drawn about the wellrepresented group accurate but the

    conclusions would not apply to the whole population. 8lthough not representa tie the data

    might be useful to see whether a hypothesi;ed relationship eDists at all that is as a test oftheory. $ata gathered from facetoface interiews for %ualitatie research hae the potential to

    yield a gold mine of insights into the peopleCs lies and situations. !here is no substitute for

    prolonged and focused conersations between trusted parties to discoer what is important to

    the interiewees and how respondents under stand key elements in their own lies. apan

    mishandled interiew data can reflect what the interiewees think the interiewers want to

    hear normatiely appropriate responsesthe fears and biases of the interiewees and the fears

    and biases of the interiewers. $ata flaws become limitations of the study for which the only

    response is to caution the reader that the results are far from certain.

    4thics and 3nstitutional Revie2 !oards

    )efore proceeding with an eDamination of research methods there are some ethical and legalconsiderations that hae obtrusiely en tered the deelopment of a research protocol. $i line

  • 7/26/2019 MIXED METHODE RESEARCH.docx

    31/42

    with designing research to be useful it should also be designed to be ethical. Paling obious

    ethical problems arise when a research procedure causes harm to those who are asked or

    forced to participate in the process. 8da seeral wellknown cases of abuse including

    psychological studies where participants were put in unusually stressful situations

    &)aumrind01:@, Milgram 017@' and medical research where participants were gien

    diseases or intentionally denied treatment &Aones 011?'. !he bureaucratic response to theseethical iolations was to create rules that would include eerybody doing any kind of research

    that inoled contact with liing people.)ureaucratic actors ealuating research they are not

    conduct ing themseles become the gatekeepers of ethi cal behaior. !his responsibility is

    misplaced, researchers themseles should be responsible for protecting the interests of

    participants in their studi. I am not naSe enough to think that all researchers are ethical or

    that institutional reiew boards &IR)s' or protection of human subjects committees will go

    away. Masalah is that ethical judgments about research hae been made eDtrinsic to the

    research process. Researchers need to design research that does just what the committees

    wantOto protect the participants of a study from harm. If researchers are not sociali;ed to

    proide these protections IR)s mi