RESEARCH METHODOLOGY
4.1 Sample Demographic
The area random sampling is used17. The population has been divided into three strata based on the geographical distribution (Jordan, Turkey and Italy). A random sample is then drawn from each of the strata. Dividing the population into a series of relevant strata means that the sample is more likely to be representative, and each of the strata is represented proportionally within the sample (figure 4.1). The total population is the sum of all the SMEs working in processing the natural stone sector in Jordan, Turkey and Italy. The total number of the firms is 2864 firms; 240 firms in Jordan, 784 firms in Turkey and 1840 firms in Italy.
Table (4.1) presents the sampling frame of Jordan, Turkey and Italy. Number of workers is the criteria for selecting the firms in the sample. The three countries (Jordan, Turkey, and Italy) follow the EU criteria; 10-250 workers.
17 The choice of sampling techniques is dependent on the feasibility and sensibility of collecting data to answer the research questions and to address the objectives from the entire population. The five main techniques are: simple random, systematic, stratified random, cluster, and multi-stage.
Figure 4.1. Population Fraction
8%
27%
65%
Jordan Turkey Italy
Table 4.1. Sampling Frame for Jordan, Turkey and Italy
# Country Population Criteria (Workers) 3 Italy 1840 10-250 Associazione Italiana
MARMOMACCHINE, 2003
The level of confidence in the estimate
The margin of error that can be tolerated
The proportion of responses that are expected to have some particular
Where n is the minimum sample size required
p% is the proportion belonging to the specified category q% is the proportion not belonging to the specified category z is the z value corresponding to the level of confidence required e% is the margin of error required
18 Factors such as the confidence that is needed in the findings, accuracy required and likely categories for analyses will affect the size of the sample that needs to be collected.
The adjusted sample size n is calculated19 as:
Where n is the adjusted minimum sample size n is the minimum sample size
N is the total population
As an example, the minimum sample for Jordan = 50% x 50% x
2
While the adjusted sample size n =
Country Population Adjusted Sample Size
N %
Jordan 240 148 140 94.5
Turkey 784 258 213 82.5
Italy 1840 320 299 93.5
Total 2864 726 652 89.8
*The estimated level of confidence = 95% (z = 1.96)
*The estimated margin or error that can be tolerated = 5%
* The estimated proportion of responses = 50% (estimated by the Jordanian Ministry of Planning in Jordan, TUMMER in Turkey and Associazione Italiana MARMOMACCHINE in Italy).
Figure (4.2) shows that the Jordanian sample presents (21%) of the total sample size, while the Turkish sample presents (33%) and the Italian sample presents (46%).
19 Because of the small total population (less than 10,000) we only need the minimum sample size.
However, this assumes a response rate of 100 percent (Saunders et al 2000).
The Chamber of Commerce in Amman provided the researcher with a list of the Jordanian companies’ names and their addresses. The interview survey was conducted by the researcher in cooperation with the Competitiveness Team in the Jordanian Ministry of Planning. By using the Statistical Package for the Social Sciences (SPSS), the sample random method has been used to select 148 SMEs. The structured interviews have been used during the period from February to March 2006. The number of the questionnaires collected is 140. The same questionnaire has been distributed to the SMEs working in the natural stone sector during the period from September 2003 to December 2003. In the first cycle the total population was 220 SMEs and only 62 questionnaires have been collected.
In Turkey, the questionnaire has been sent to 258 SMEs by fax or via e-mail in September 2004. The Turkish Association of Marble, Natural Stone and Machinery Products (TUMMER), Izmir Marble Producers Association, and the Turkish Embassy in Tel Aviv, Israel, provided the names and addresses of the companies to the researcher.
Many telephone calls and reminder faxes have been sent to these firms. Unfortunately, by the end of the year few firms returned the questionnaires back. Two university
business-Figure 4.2. Sample Fraction (Received)
21%
33%
46% Jordan
Turkey Italy
students20 from Ankara were asked by the researcher to follow up and collect the questionnaires. By the end of December 2005, 213 questionnaires were collected. The researcher visited different associations during his visit to Turkey from 28/3/2004 till 10/4/2004. In addition, the researcher attended the International Marble Fair in Izmir during that period and visited different associations as mentioned in table (5.5).
In Turkey, almost (95%) of the natural stone industry is concentrated in Marmara, Aegean regions, Mediterranean, Central Western Anatolia and Central Anatolia.
According to TUMMER (2004), there are 2500 operations in Turkey and almost 800 firms are small to medium sized enterprises and the rest are either large-scale enterprises or workshops. The random sample method is used to select 258 firms out of 784. The survey covered different provinces such as Balikesir, Bursa, Kirklareli, Afyon, Aydin, Izmir, Mulga, Kutahya, Ankara, Eskisehir, Kirsehir and Nigde, Adana, and Burdur.
In Italy, the questionnaire has been sent by fax or via the e-mail to 320 SMEs during September 2004. These firms have been selected randomly by using the SPSS. Names and addresses of these firms have been prepared with the help of the Associazione Italiana MARMOMACCHINE, Rome Chamber of Commerce, and Italian Consulate in Jerusalem, Israel. Many telephone calls and reminder faxes have been sent to these firms.
Few firms returned back the questionnaires by the end of the year. The researcher asked a Palestinian businessman, who holds a BA in Business Administration21 and owns a representative office for stone and marble in Verona, Italy, to follow up and collect back the questionnaires. By the end of December 2005, 299 questionnaires were returned back.
During May 2004, the researcher managed to visit different associations in Italy that are listed in table (5.5).
20 The two university students were working under the supervision of Ms. Sercil Tokoglu who holds an MBA from Maastricht School of Management, the Netherlands in 1997.
21 The name of the businessman is Mr. Fahed Gaith. Mr. Gaith was an undergraduate student in Hebron University where the researcher was teaching as a part-timer.