3. Research methodology and samplin

3. Research methodology and sampling procedure
Section:Previous sectionNext section
3.1. Model specification and measurement of variables
Equation (1) is modeled to test H1a, H1b and H1c, i.e. to study the effect of family ownership on dividend payout:

(Equation 1)

The dependent variable, dividend payout ratio (DPR) is measured as the total amount of annual common dividends paid out, scaled by the net income after tax. Share repurchases or share buy-backs are not examined in this study due to their low occurrence in developing markets, in line with the approach of La Porta et al. (2000). For firm-years with negative DPR, the value “0” (zero) is assigned. Truncating dividend at zero thus justifies the use of Tobit regression analysis. Firm-specific control variables included in the model are profitability (ROE; net income after tax as a percentage of shareholder’s fund), DEBT (long-term debt as a percentage of total assets), future investment/growth opportunities (GROWTH OPP; total value of market capitalization plus total book value of liabilities minus book value of common equity divided by total assets), current and past growth (PAST GROWTH; annual growth in sales) and firm size (SIZE; natural log of sales) as these variables have been found in prior studies to be statistically significant in influencing dividend payout. DEBT, GROWTH OPP and PAST GROWTH have been documented to be endogenous with dividend payout (Jensen et al., 1992; Zhang, 2008)[4]. Hence, in an effort to mitigate these endogeneity concerns, the lagged one-year values of these variables are used in this study (Hill et al., 2010). However the limitation of these estimations in that it is not possible to fully eliminate the issue of endogenity is acknowledged. Consistent with Setia-Atmaja et al. (2009), controls for board governance characteristics include board size (BD_SIZE; natural log of the total number of directors on the board) and board independence (BD_IND; fraction of independent non-executive directors on the board of directors).

Chen et al. (2005) define family ownership using three variables based on cut-off ownership levels identified by reference to the economic and legal framework of the Hong Kong market and due to its non-linear relationship with dividend payout. The use of cut-off ownership levels allows the examination of the agency conflict between family owners and minority shareholders via dividend payout in greater depth, as opposed to the use of a single overall measure of a continuous variable representing family ownership. Family share ownership (PFO) is measured as the fraction of total company shares held by the family. Family ownership is divided using three cut-off levels. First, a 5 percent cut-off is used as this is the minimum ownership level that requires shareholders to disclose their shareholdings and interest in investee companies in Malaysia[5]. Next, a 33 percent cut-off is used as it triggers a mandatory public offer when shareholders hold a minimum of 33 percent of shares of a company in Malaysia. Accordingly, the variable PFO_LESS5 represents dispersed family ownership at between 0 and 5 percent, PFO_5TO33 represents fairly concentrated family ownership at between 5 and 33 percent, while PFO_MORE33 represents highly concentrated family ownership at greater than 33 percent. Assuming actual family ownership fraction to be m:

(Equation 2)

(Equation 3)

(Equation 4)

The measurements of all the variables are also presented in Table I.

Hasil (Bahasa Indonesia) 1:

3. penelitian metodologi dan sampling prosedurBagian: bagian sectionNext sebelumnya3.1. model spesifikasi dan pengukuran dari variabelPersamaan (1) adalah model untuk menguji H1a, H1b dan H1c, yaitu untuk mempelajari efek dari kepemilikan keluarga di deviden:(Persamaan 1)The dependent variable, dividend payout ratio (DPR) is measured as the total amount of annual common dividends paid out, scaled by the net income after tax. Share repurchases or share buy-backs are not examined in this study due to their low occurrence in developing markets, in line with the approach of La Porta et al. (2000). For firm-years with negative DPR, the value “0” (zero) is assigned. Truncating dividend at zero thus justifies the use of Tobit regression analysis. Firm-specific control variables included in the model are profitability (ROE; net income after tax as a percentage of shareholder’s fund), DEBT (long-term debt as a percentage of total assets), future investment/growth opportunities (GROWTH OPP; total value of market capitalization plus total book value of liabilities minus book value of common equity divided by total assets), current and past growth (PAST GROWTH; annual growth in sales) and firm size (SIZE; natural log of sales) as these variables have been found in prior studies to be statistically significant in influencing dividend payout. DEBT, GROWTH OPP and PAST GROWTH have been documented to be endogenous with dividend payout (Jensen et al., 1992; Zhang, 2008)[4]. Hence, in an effort to mitigate these endogeneity concerns, the lagged one-year values of these variables are used in this study (Hill et al., 2010). However the limitation of these estimations in that it is not possible to fully eliminate the issue of endogenity is acknowledged. Consistent with Setia-Atmaja et al. (2009), controls for board governance characteristics include board size (BD_SIZE; natural log of the total number of directors on the board) and board independence (BD_IND; fraction of independent non-executive directors on the board of directors).Chen et al. (2005) define family ownership using three variables based on cut-off ownership levels identified by reference to the economic and legal framework of the Hong Kong market and due to its non-linear relationship with dividend payout. The use of cut-off ownership levels allows the examination of the agency conflict between family owners and minority shareholders via dividend payout in greater depth, as opposed to the use of a single overall measure of a continuous variable representing family ownership. Family share ownership (PFO) is measured as the fraction of total company shares held by the family. Family ownership is divided using three cut-off levels. First, a 5 percent cut-off is used as this is the minimum ownership level that requires shareholders to disclose their shareholdings and interest in investee companies in Malaysia[5]. Next, a 33 percent cut-off is used as it triggers a mandatory public offer when shareholders hold a minimum of 33 percent of shares of a company in Malaysia. Accordingly, the variable PFO_LESS5 represents dispersed family ownership at between 0 and 5 percent, PFO_5TO33 represents fairly concentrated family ownership at between 5 and 33 percent, while PFO_MORE33 represents highly concentrated family ownership at greater than 33 percent. Assuming actual family ownership fraction to be m:(Equation 2)(Equation 3)(Equation 4)The measurements of all the variables are also presented in Table I.

Hasil (Bahasa Indonesia) 2:

3. Penelitian metodologi dan prosedur pengambilan sampel
Bagian: Sebelumnya Bagian Bagian
3.1. Spesifikasi Model dan pengukuran variabel
Persamaan (1) adalah model untuk menguji H1a, H1b dan H1c, yaitu untuk mengetahui pengaruh kepemilikan keluarga pada dividend payout:

(Persamaan 1)

Variabel terikat, rasio pembayaran dividen (DPR) diukur sebagai jumlah total dividen umum tahunan dibayarkan, skala oleh laba bersih setelah pajak. Pembelian kembali saham atau share buy-back tidak diteliti dalam penelitian ini karena terjadinya rendah di pasar berkembang, sejalan dengan pendekatan La Porta et al. (2000). Untuk perusahaan-tahun dengan DPR negatif, nilai "0" (nol) ditugaskan. Truncating dividen nol sehingga membenarkan penggunaan analisis regresi Tobit. Variabel kontrol khusus perusahaan termasuk dalam model yang profitabilitas (ROE; laba bersih setelah pajak sebagai persentase dari dana pemegang saham), (utang jangka panjang sebagai persentase dari total aset) UTANG, peluang investasi / pertumbuhan di masa depan (PERTUMBUHAN OPP; Total nilai kapitalisasi pasar ditambah nilai buku dari kewajiban dikurangi nilai buku ekuitas umum dibagi dengan total aset), saat ini dan pertumbuhan masa lalu (pERTUMBUHAN mASA lALU, pertumbuhan tahunan penjualan) dan ukuran perusahaan (sIZE; log natural dari penjualan) sebagai variabel ini memiliki telah ditemukan dalam studi sebelumnya yang signifikan secara statistik dalam mempengaruhi dividen payout. UTANG, PERTUMBUHAN OPP dan PERTUMBUHAN MASA LALU telah didokumentasikan untuk menjadi endogen dengan dividend payout (Jensen et al, 1992;. Zhang, 2008) [4]. Oleh karena itu, dalam upaya untuk mengurangi kekhawatiran endogeneity ini, tertinggal nilai satu tahun yang variabel ini digunakan dalam penelitian ini (Hill et al., 2010). Namun keterbatasan estimasi ini di bahwa itu tidak mungkin untuk sepenuhnya menghilangkan masalah endogenity diakui. Konsisten dengan Setia-Atmaja et al. (2009), kontrol untuk karakteristik pemerintahan papan termasuk ukuran papan (BD_SIZE; log alami dari jumlah direksi di papan) dan independensi dewan (BD_IND; fraksi direktur non-eksekutif independen pada dewan direksi).

Chen et al. (2005) mendefinisikan kepemilikan keluarga menggunakan tiga variabel berdasarkan tingkat kepemilikan cut-off diidentifikasi dengan mengacu pada kerangka kerja ekonomi dan hukum dari pasar Hong Kong dan karena hubungan non-linear dengan dividen payout. Penggunaan tingkat kepemilikan cut-off memungkinkan pemeriksaan konflik keagenan antara pemilik keluarga dan pemegang saham minoritas melalui dividen payout secara lebih mendalam, yang bertentangan dengan penggunaan ukuran keseluruhan tunggal dari variabel kontinu mewakili kepemilikan keluarga. Kepemilikan saham keluarga (PFO) diukur sebagai fraksi dari total saham perusahaan yang dimiliki oleh keluarga. Kepemilikan keluarga dibagi menggunakan tiga tingkat cut-off. Pertama, 5 persen cut-off digunakan karena ini adalah tingkat kepemilikan minimum yang mengharuskan pemegang saham untuk mengungkapkan kepemilikan saham dan minat mereka dalam perusahaan asosiasi di Malaysia [5]. Selanjutnya, 33 persen cut-off digunakan sebagai memicu penawaran umum wajib ketika pemegang saham memegang minimal 33 persen saham dari sebuah perusahaan di Malaysia. Dengan demikian, variabel PFO_LESS5 mewakili kepemilikan keluarga yang tersebar di antara 0 dan 5 persen, PFO_5TO33 mewakili kepemilikan keluarga cukup terkonsentrasi di antara 5 dan 33 persen, sementara PFO_MORE33 mewakili kepemilikan keluarga yang sangat terkonsentrasi di lebih dari 33 persen. Dengan asumsi fraksi kepemilikan keluarga yang sebenarnya menjadi m:

(Persamaan 2)

(Persamaan 3)

(Persamaan 4)

Pengukuran semua variabel juga disajikan dalam Tabel I.