Sampling Procedure
Stratified random sampling method was used to select the sample.
Three levels of stratification were used: industry, establishment size, and region.
Industry stratification was designed in the way that follows: the universe was stratified into four manufacturing industries and four services industries- Food (NACE 2.0 codes 10), Textiles and Wearing apparel (NACE codes 13 and 14), Fabricated Metal, Machinery and Motor Vehicles (NACE codes 25, 28, 29), Other Manufacturing (NACE codes 11, 12, 15-24, 26, 27, 30-32), Construction (NACE codes 41, 42, 43) Wholesale and Retail (45, 46, 47), Transport (49, 50-53) and Other Services (NACE codes 33, 55, 56, 58, 61, 62, 79, and 95).
Siize stratification was defined as follows: micro (less than 5 employees), small (5 to 19 employees), medium (20 to 99 employees), and large (100 or more employees).
Regional stratification was done including eighty-one NUTS 3 regions: Istanbul, Tekirdag , Edirne, Kirklareli, Balikesir, Çanakkale, Izmir, Aydin, Denizli, Mugla, Manisa, Afyonkarahisar, Kütahya, Usak, Bursa, Eskisehir, Bilecik, Kocaeli, Sakarya, Düzce, Bolu, Yalova, Ankara, Konya, Karaman, Antalya, Isparta, Burdur, Adana, Mersin, Hatay, Kahramanmaras, Osmaniye, Kirikkale, Aksaray, Nigde, Nevsehir, Kirsehir, Kayseri, Sivas, Yozgat, Zonguldak, Karabük, Bartin, Kastamonu, Çankiri, Sinop, Samsun, Tokat, Çorum, Amasya, Trabzon, Ordu, Giresun, Rize, Artvin, Gümüshane, Erzurum, Erzincan, Bayburt, Agri, Kars, Igdir, Ardahan, Malatya, Elazig, Bingöl, Tunceli, Van, Mus, Bitlis, Hakkâri, Gaziantep, Adiyaman, Kilis, Sanliurfa, Diyarbakir, Mardin, Batman, Sirnak, and Siirt. However, for precision purposes, the minimum number of observations required for a minimum level of precision of 7.5% of estimates of proportions with a 95% confidence interval was guaranteed for these regions grouped into 26 NUTS 2 regions: TR10, TR21, TR22, TR31, TR32, TR33, TR41, TR42, TR51, TR52, TR61, R62, TR63, TR71, TR72, TR81, TR82, TR83, TR90, TRA1, TRA2, TRB1, TRB2, TRC1, TRC2, TRC3.
The sample frame consisted of a listing of firms from the Turkish Statistical Institute (TUIK). The quality of the frame was enhanced by the verification process conducted by the three contractors. However, the sample frame was not immune from the typical problems found in establishment surveys: positive rates of non-eligibility, repetition, non-existent units, etc.
Given the impact that non-eligible units included in the sample universe may have on the results, adjustments are needed when computing the appropriate weights for individual observations. The percentage of confirmed non-eligible units as a proportion of the total number of sampled establishments contacted for the survey was 3.8% (1,646 out of 43,421 establishments).