 |
|
|
|
||||||||||
|
|||||||||||||
|
|
|
|
||||||||||||||||||||
|
|||||||||||||||||||||||
|
Click here to download a file version of the Overview Of Services page. |
|||||||||||||||||||||||
|
Overview Of Services Background Scientific Telephone Samples (STS) is one of the largest and most respected sampling suppliers in the United States. The company was founded in 1988, and currently serves as a supplier of random digit (RDD) samples and listed samples to over 500 major marketing research firms, client companies and university-affiliated survey research centers. We use the highest quality, state-of-the-art procedures to construct random digit and listed samples. Clients can rest assured our added dimension of broad-based research knowledge will help guarantee every sample developed by STS will be the finest available anywhere. STS is proud to have been a driving force for many of the changes in the fundamental structure of the sampling business during the past several years. We were the first to offer same-day service, a standard 24-hour turnaround, a "security blanket" of 10% extra numbers free, free record coding, the choice of multiple call record sheets, the use of ACORN geo-demographic segmentation data for targeted RDD samples, the marking of households from the National Telephone Preference List of people objecting to unsolicited telephone calls, and a host of other innovations. Importantly, STS was the first company to adopt an aggressive value-oriented pricing philosophy and continues to be the market leader with regard to cost efficiency. Our standard price for RDD samples ranges from 2¢ to 5¢ per number, depending on yearly volume, size of order, or contractual agreement. Overall, STS is a prime example of the ability to use state-of-the-art computer technology and database integration techniques as a means of achieving significant innovation and cost reduction. STS operates in a network environment using the most advanced computer technology. The daily output of this system is virtually unlimited.
Types Of Samples Our sampling frame is based on the largest database of working residential telephone exchanges and working blocks in the United States. Thus, we can draw from any part of the U.S.A. — no matter how large or small. The information is updated several times per year, and cross-checked monthly against area code and assigned exchange lists furnished by the telephone companies. Exchanges and/or working blocks designated for business or governmental telephones, car/boat/plane mobile units, and other commercial or institutional services, are screened out. STS can furnish any type of random digit sample desired, including:
Planning Your Sample An STS representative is always available to assist you in developing a sampling plan for your project. Our specialists have formal marketing research experience from either the client or supplier side, and understand your needs from a technical research standpoint. To help with sample design, STS maintains a complete sampling library, including the latest census data, an in-house mapping system, a full complement of computer databases, and our own detailed market analyses.
Determining How Much Sample To Buy We will be pleased to help you decide how many numbers will be needed. There are four major considerations to deal with:
The appendix of this manual includes a chart listing the quantities suggested by two of the more popular "formulas" for determining how much sample to buy.
Selection Methodology Scientific Telephone Samples can produce any type of random digit sample, utilizing all of the standard sampling methodological selection methods: Type A(EPSEM), Type B(weighted), cluster, Mitofsky Waksberg, or even Sudman. Most RDD samples are constructed using one of two different methodologies for pulling working blocks, either Type A (EPSEM /pure/unweighted) or Type B (weighted/efficient). Either type can be used regardless of the geographic sampling unit (i.e. state, county, ZIP, etc.). Type A samples are pulled using a strict definition of randomness. They are called "unweighted" samples because each possible telephone number within the appropriate working blocks has an equal chance of being selected to generate a random digit number. Completed interviews from a Type A sample which has been dialed to exhaustion should be highly representative of the population under study. Type B or "weighted" samples are pre-weighted so RDD numbers are created from telephone working blocks in proportion to the number of estimated household listings in each working block. Working blocks which are more densely packed with numbers will be more prevalent in your sample. For example, a working block which has 50 known numbers in existence would have twice the probability of being included as one which had just 25 numbers. Type B samples are most useful when one is willing to overlook a strict definition of randomness in favor of slightly greater calling efficiency, because of fewer "disconnects." In theory, completed interviews from Type B samples may tend to over-represent certain types of working blocks, but many researchers feel there is not much difference in representativeness.
Replicates Unless otherwise instructed, STS samples are constructed in "replicates." A replicate is a representative subset of the total sample. Dividing your sample into replicates reduces potential bias associated with the way the sample is administered. Each replicate is also randomly shuffled within itself, so you do not need to randomize the sample in any way. You may specify any number of replicates desired. For details on replicate administration and setting replicate size, see the Appendix.
Record Coding Unlike some sampling suppliers, STS does not charge extra for record coding. When received in a computer file, RDD phone numbers can be appended with as many geo-demographic record codes as you wish. The most useful record codes are typically ZIP codes and FIPS (county) codes. Other useful codes such as city name, median income, racial breakdowns, % without telephone, age, and ACORN data can aid in sample control. There are nearly 200 geo-demographic and lifestyle codes from which to choose. Remember that the record codes shown for RDD samples are simply indicative of the areas covered by the sample. For large areas such as counties, MSA's, DMA's, ADI's, or states, the record coding by FIPS codes is highly accurate. However, record coding by ZIP code should not be considered as precise. This is true for any RDD sample since telephone working blocks (NPA-NXX-XX) cross over ZIP code boundaries. If you must know the true location of the household — particularly within small areas such as a ZIP code or city — you must clarify with each respondent.
Demographic Targeting You can also "target" RDD samples to improve calling efficiencies by including or excluding ZIP codes based on demographic criteria. It is important to be mindful of the fact that restricting an RDD sample only to those areas which match pre-selected demographic criteria destroys the true representativeness of the sample. On the other hand, geographic stratification can be used to eliminate a good deal of this bias. Moreover, many researchers feel targeted RDD samples are still a better alternative than using a listed sample. In addition, commercially prepared listed samples have their own types of biases — the fact listed samples do not include households with unlisted telephone numbers can be a severe methodological problem. In many major metro areas, unlisted rates are in the 40% to 60% range. Some of the more commonly requested inclusions/exclusions are:
Altogether, STS offers nearly 200 demographic, lifestyle, and household variables for targeting samples.
Output Format STS can output your sample in almost any format needed, including:
STS samples shipped on disk, via e-mail, or via modem are fully compatible with all CATI systems, such as Ci2, Ci3, ACS-Query, CMFC, Athena, etc. The sample files are usually sent in a fixed field ASCII format, which is a widely accepted CATI standard, but can also be configured tab delimited, comma delimited, or in virtually any format requested.
Pre-IDentification Of Disconnects: STS PreID Disconnects have traditionally been one of the most inefficient parts of random digit sampling. Unfortunately, the cost to eliminate these disconnects prior to being "in the field" has been relatively high. With STS PreID, maximizing efficiency by pre-identifying disconnects doesn't have to compromise your budget. Highly accurate, fast, and inexpensive, STS PreID is the state-of-the-art in disconnect purging. STS PreID will normally identify 50% to 75% of all disconnects in any RDD sample. Many fax numbers are also identified. STS PreID disconnect purging requires only a minimal amount of additional time -- typically 24 hours or less, depending on the number of records. STS PreID costs just 1¢ to 1.5¢ per number depending on annual volume.
Pricing You will find STS to be the low price leader in the sampling industry. Our standard base price for full service random digit samples is just 5¢ per number, and we have an aggressive discount schedule so that frequent buyers can pay as little as 3¢ each. Additional discounts may be earned for large orders. For larger accounts, a contract may be signed to provide even better discounts down to 2¢ per number. (See separate price schedule for exact details. Prices subject to change without notice.)
Timing STS has the fastest turnaround in the business. Because of our advanced computer technology, STS always ships or modems full service RDD samples the very next business day (by 12 P.M. PST) at no extra cost. And, for just one additional cent per number, you can guarantee your sample will be shipped or sent electronically the same day it is ordered. Same-day service orders can be ready in as little as 30 minutes or less.
|
|||||||||||||||||||||||
