Description of Sampling Procedures for 1997, 2000, 2007 Waves of nc1030 Data Description of the Sampling Procedures for the 1997 Data

Description of the Sampling Procedures for the 2007 Data

Download 24.3 Kb.
Size24.3 Kb.
1   2   3
Description of the Sampling Procedures for the 2007 Data
Survey Procedures: The NFBS panel data are longitudinal over ten years with the third wave containing experiences with natural disasters. Such data allows us to distinguish between business failures due to a natural disaster and those failures that are “routine” business failures. Only a small body of disaster research exists on individual responses that estimate disaster prevalence, nature of losses, and predictors of recovery. These studies have been conducted either in urban disaster locations or with small convenience samples. All samples have been cross sectional. Studying businesses at one point in time potentially has presented a distorted picture, since businesses may fail at a later date; economic cycles and increased indebtedness after the disaster can influence success and failure over the long-term (Tierney, 1997). The NFBS panel data can begin to provide some insight into the distinction in failure rates.
In the mid-1990’s, a group of academic scholars from seventeen universities combined their resources to form the Family Business Research Group and conduct research on the impact and interplay between owning/managing a business and managing a family. Research began with national household screening to locate people with a family in the home who owned and managed a business, and interviews were conducted with both business managers and household managers in 1997 (Winter, Fitzgerald, Heck, Haynes & Danes, 1009). Follow-up interviews were conducted with 1997 respondents three years later, in 2000 (Winter, Danes, Koh, Fredericks & Paul, 2004), and ten years later in 2007. The 2007 study was designed to obtain updated information about the status of the businesses and families and also to learn about the impact of natural disasters and disaster relief on business viability. Data collection for each segment of the research was conducted by Iowa State University’s Center for Survey Statistics and Methodology (CSSM), formerly known as the Statistical Laboratory.
The sample for the 2007 survey consisted of the 708 cases that were attempted for the 2000 study. One of the 1997 Combination Managers had died prior to the 2000 study. Efforts were made to locate and interview all of the remaining 707 cases, including those that were unlocatable, refusals, or closed out after a maximum number of attempts in 2000. Cases interviewed in 2000 with businesses open or closed, sold or owned, would be contacted again and efforts would be made to interview the same respondents who were interviewed in 2000.
Participants in the 1997 and 2000 surveys had received project newsletters from the Family Business Research Group periodically to maintain participant interest and accurate contact information. However, no newsletters had been sent for several years. Because of this, efforts were made to verify the address and telephone number of the sample, and a newsletter was sent to the sample prior to any interviewing attempts in order to make use of postal address updates.
The earlier surveys did not include a monetary incentive for respondents. This time, however, it was decided to incorporate an incentive into the study design. If it was known in advance that a business was closed or sold, the interview was comparatively brief and those cases were sent $10 after survey completion to thank them for their time. If, however, the business status was unknown in advance and the potential interview length was possibly as long as 40 minutes, those cases were sent $20 after survey completion.
Possible contact information was available for 679 of the 707 cases in the sample. After the newsletter distribution and various tracking efforts, reliable addresses were available for 574 of the 707 cases. Data collection took place from May 9 to August 6, 2007. Initial telephone screening was conducted using paper documents. Interviews were conducted using computer-assisted telephone interviewing (CATI) Blaise software. Phone numbers with no personal contact were rotated through a minimum of 15 call attempts at various times (e.g., days and evenings, weekdays and weekends).
Of the 708 cases, 29.4% could not be located, and 22.2% resulted in either refusals or maximum calls classifications with no screening completed. Most cases that completed the initial screening (verifying the business status) also completed interviews. Of the 334 cases screened in 2007, 290 completed short or long interviews for a Cooperation Rate of 86.8%. Response rates can be viewed in several ways. The 290 cases with completed short or long are 86.8% of the 334 cases that completed initial screening, 58.6% of the 495 eligible cases that could be located, and 41.0% of the 708 total cases.

Share with your friends:
1   2   3

The database is protected by copyright © 2020
send message

    Main page