Absent relevant archival data, rese

Absent relevant archival data, researchers may use experimental tests to study real-
world phenomena. However, it can be difficult to identify appropriate participants for the
experiments. We chose to use graduate business students who were novices in the use of
the BSC. Other potential subject groups include employees of a particular BSC firm, em-
ployees in a cross-section of BSC firms, or people who have been trained in the use of a
BSC. These different subject pools have different advantages and disadvantages, and theory
does not suggest an optimal choice. In a study of the use of common vs. unique measures,
focusing on employees of a particular BSC firm as they use their own actual scorecards
would compromise experimental control. Specifically, performance on common measures
would need to differ substantially from performance on unique measures in order to deter-
mine the effects of the two types of measures. Using employees from a cross-section of
BSC firms means that most or all participants would be presented with scorecards different
from those used in their own firms and units. In this case, participants’ own experiences
with specific measures may differ from the situation presented in the experimental materials
(e.g., catalog profits is a unique measure in our experiment, but it could be a common
measure in some firms). This might lead a participant to weight a measure based on his or
her experience with the measure (e.g., high weight for a measure that is common in the
participant’s firm) rather than responding to the manipulated level of the measure (e.g.,
unique in the experimental instrument) so that test results may not capture the effects of
manipulated variables. Finally, training experimental participants to use a scorecard can
lead to significant demand effects. For example, if we trained our student-subjects to weight
BSC items in a particular way or to use a particular strategy for interpreting the scorecards
(e.g., using a causal diagram linking the measures to one another), our tests of the use of
measures would largely reflect the success of our training program. Thus, although it is
tempting to use subjects with BSC experience, this presents difficulties for a controlled and
generalizable study.

Hasil (Bahasa Indonesia) 1:

Absen data arsip yang relevan, peneliti dapat menggunakan tes eksperimental untuk mempelajari real-fenomena dunia. Namun, itu bisa sulit untuk mengidentifikasi sesuai pesertapercobaan. Kami memilih untuk menggunakan mahasiswa pascasarjana bisnis yang pemula dalam penggunaan dariBSC. Kelompok mata pelajaran lain potensi termasuk karyawan tertentu BSC perusahaan, em-ployees dalam keratan BSC perusahaan, atau orang-orang yang telah dilatih dalam penggunaanBSC. Kolam berbeda subjek ini memiliki berbagai keuntungan dan kerugian, dan teoritidak menyarankan pilihan yang optimal. Dalam sebuah studi dari penggunaan umum vs langkah-langkah yang unik,berfokus pada karyawan perusahaan BSC tertentu karena mereka menggunakan mereka sendiri sebenarnya Scorecardakan membahayakan eksperimental kontrol. Secara khusus, kinerja pada langkah-langkah umumperlu berbeda secara substansial dari kinerja pada langkah-langkah yang unik untuk menghalangi-tambang efek dari dua jenis tindakan. Menggunakan karyawan dari penampangBSC perusahaan berarti bahwa sebagian atau semua peserta akan disajikan dengan Scorecard berbedadari mereka yang digunakan dalam perusahaan mereka sendiri dan unit. Dalam kasus ini, para peserta sendiri pengalamandengan ukuran tertentu mungkin berbeda dari situasi yang disajikan dalam bahan eksperimental(misalnya, Katalog keuntungan adalah ukuran yang unik dalam percobaan kami, tetapi bisa menjadi umummengukur di beberapa perusahaan). Ini mungkin menyebabkan peserta berat ukuran berdasarkan nya atauher experience with the measure (e.g., high weight for a measure that is common in theparticipant’s firm) rather than responding to the manipulated level of the measure (e.g.,unique in the experimental instrument) so that test results may not capture the effects ofmanipulated variables. Finally, training experimental participants to use a scorecard canlead to significant demand effects. For example, if we trained our student-subjects to weightBSC items in a particular way or to use a particular strategy for interpreting the scorecards(e.g., using a causal diagram linking the measures to one another), our tests of the use ofmeasures would largely reflect the success of our training program. Thus, although it istempting to use subjects with BSC experience, this presents difficulties for a controlled andgeneralizable study.