Return to the Table of Contents
Return to Appendix 1

 

Introduction

 

 

All the secondary sources used in the project provide extensive documentation and their web sites are supplied on at the beginning of Appendix I of this Needs Assessment report.

Our own primary data collection included the qualitative components (key informant and focus group interviews) and the household survey conducted by PRC, Inc.

The methodology for the survey is described in Appendix II and the whole survey instrument is supplied in Appendix II. The methodology was essentially identical (except for over sampling Upper Bucks and low income areas) to that used generally by the CDC’s Behavioral Risk Factor Survey that is conducted annually for Pennsylvania as a whole by the State Health Department. The response rate, a key measure of the quality of the survey, was somewhat better for our survey than that which was achieved by the State’s most recent statewide survey. The data can be accessed and further analyzed at : www.PRCEasyView.com. (user name = Bucks and password = Survey).

 

 

1. METHODOLOGY

A precise and carefully executed methodology is critical in asserting the validity of the results gathered in the 2001 Bucks County Health & Quality of Life Survey. The following will describe the various aspects of the sample design and data collection utilized for this study.

Survey Methodology

To ensure the best representation of the population surveyed, a telephone interview methodology was employed. The primary advantages of telephone interviewing are timeliness, efficiency, and random selection capabilities. Keep in mind that the Federal Communications Commission (FCC) estimated in 2000 that only 3.4% of Pennsylvania households do not have telephones.

Sample Design

The sample design utilized for this research effort consisted of a random sample of 900 households in Bucks County, using a stratified sample equally distributed among three sub-county regions: Upper Bucks, Central Bucks and Lower Bucks County (described in the following map). This stratification ensures a comparable maximum error rate associated with each of the three regions — findings within each sub-county region carry a maximum error rate of ±5.7% at the 95% confidence level. The maximum error rate associated with the total sample of interviews is ±3.1% at the 95% confidence level.

The random sample of consumer households was selected in proportion to the actual population distribution of each region (at the ZIP Code level) so that a sample representative of each area was obtained. Population estimates were based on 2000 census estimates of adults aged 18 and over as published by CACI, Inc.

Furthermore, in order to ensure representation among lower income households, an over-sample of an additional 101 interviews was conducted among households likely to be low income; this sample was obtained from purchased lists of telephone numbers for households with categorized incomes below $20,000.

A total of 61 interviews were achieved among households with annual incomes under $15,000; 112 interviews were achieved among households with annual incomes under $20,500; 142 were achieved among households below the 200% poverty threshold.

Once all of the interviewing was completed, the sample was weighted by key demographic characteristics and in proportion to the population at the ZIP Code level countywide, so that a representative sample was obtained.

The following table illustrates the population distribution for the defined area, accompanied by the number of interviews targeted and achieved for each ZIP Code.

Note: for ZIP Codes that cross county lines, interviews were only completed with households within Bucks County; population figures were adjusted accordingly.

 

 

 

Area

ZIP Code

2000 Estimated Adult Population

Actual Random Interviews

Oversample Interviews

Weighted Sample

Upper Bucks

18036- Coopersburg

943

4

9

2

* 18081- Springtown

*

*

*

*

18041- East Greenville

0

1

1

0

18054- Green Lane

293

4

0

1

18073- Pennsburg

214

3

1

0

18076- Red Hill

1558

0

0

0

18077- Riegelsville

1041

4

0

2

* 18039- Durham

*

*

*

*

18917- Dublin

1370

3

0

3

18920- Erwinna

384

1

0

1

18927- Hilltown

33

0

0

0

18930- Kintnersville

2016

9

2

5

* 18921- Ferndale

*

*

*

*

18942- Ottsville

2407

10

1

5

* 18910- Bedminster

*

*

*

*

* 18953- Revere

*

*

*

*

18944- Perkasie

14815

69

4

34

* 18911- Blooming Glen

*

*

*

*

* 18962- Silverdale

*

*

*

*

18951- Quakertown

23630

110

9

54

* 18935- Milford Square

*

*

*

*

* 18968- Spinnerstown

*

*

*

*

* 18970- Trumbauersville

*

*

*

*

* 18981- Zionhill

*

*

*

*

18955- Richlandtown

*

*

*

*

18960- Sellersville

8280

35

3

19

18964- Souderton

315

5

1

1

18969- Telford

1844

23

3

4

18972- Upper Black Eddy

2893

11

2

7

19440- Hatfield

655

8

0

1

Area Total:

62691

300

36

139

 

 

Central Bucks

ZIP Code

2000 Estimated Population

Actual Random Interviews

Oversample Interviews

Weighted Sample

18901- Doylestown

32798

68

11

74

* 18912- Buckingham

*

*

*

*

* 18916- Danboro

*

*

*

*

18913- Carversville

35

0

0

18914- Chalfont

12427

25

4

28

18923- Fountainville

169

0

0

0

18925- Furlong

3054

6

2

7

* 18922- Forest Grove

*

*

*

*

* 18956- Rushland

*

*

*

*

* 18980- Wycombe

*

*

*

*

18929- Jamison

4151

8

9

18932- Line Lexington

190

0

0

0

18933- Lumberville

139

0

0

0

18934- Mechanicsville

134

0

0

0

18938- New Hope

9210

19

4

21

* 18928- Holicong

*

*

*

*

* 18931- Lahaska

*

*

*

*

* 18946- Pineville

*

*

*

*

* 18963- Solebury

*

*

*

*

18947- Pipersville

3718

8

0

8

* 18926- Gardenville

*

*

*

*

* 18949- Plumsteadville

*

*

*

*

* 18950- Point Pleasant

*

*

*

*

18954- Richboro

6840

14

0

16

18966- Southampton

31453

64

0

71

18974- Warminster

30509

62

3

69

18976- Warrington

10014

20

3

23

18977- Washington Crossing

2757

6

1

6

Area Total:

147598

300

28

332

Lower Bucks

18940- Newtown

18969

24

2

43

* 18943- Penns Park

*

*

*

*

19007- Bristol

17313

22

2

39

19020- Bensalem

40782

53

9

93

19021- Croydon

7803

10

1

18

19030- Fairless Hills

10455

13

3

24

19047- Langhorne

23211

30

3

53

19053- Feasterville Trevose

21074

27

1

48

19054- Levittown

12816

16

3

29

19055- Levittown

11450

15

4

26

* 19058- Levittown

*

*

*

*

19056- Levittown

14365

18

3

33

19057- Levittown

13701

18

1

31

19067- Morrisville

41120

54

5

93

Area Total:

233059

300

37

530

GRAND TOTAL:

443348

900

101

1001

* Denotes a Post Office Box ZIP Code, rather than a geographical area of residence. For the purposes of this study, persons with a P.O. Box address would instead be categorized within the geographic ZIP Code in which they actually live (this may or may not be as these PO Box ZIP Codes are categorized above).

Sampling Technique

For the random surveys, the sampling used for this study consisted of a two-stage, modified Waksberg technique. Two distinct levels of attention were given to the data collection, as follows:

STAGE 1: The primary sampling unit (PSU) included in the survey consisted of the first eight numbers of each 10-digit telephone number (i.e., area code + prefix + next two digits). These PSUs were selected randomly from a listing of valid telephone numbers for the defined area obtained from Donnelley Services.

Next, two randomly generated digits were added to the PSU to produce a 10-digit telephone number to be contacted. Thus, all households in the defined area share equal opportunity to be interviewed. At least five attempts were made to complete an interview with each PSU.

STAGE 2: At the time of interview, individual respondents were selected randomly at the household level according to age ranking. Interviewers first determined the number of household members aged 18 and over, then a random digit key instructed them which household member to interview (based on age ranking). Proportionate numbers of interviews were conducted with the oldest members of households, the second oldest, the third oldest, and so on. If this household member was not available, appointments were set to speak with the desired respondent. This way, an accurate sampling of the population was obtained, one which does not bias toward specific age groupings, decision-makers or heads of households.

All administration of the surveys, data collection, and data processing was conducted by Professional Research Consultants, Inc. (PRC), a nationally recognized research firm specializing in the health care field. PRC's experienced staff of professional interviewers conducted the interviews. Fifteen percent of all interviews were verified by PRC's supervisors, and several quality control procedures were used to ensure accurate coding and data integrity.

Response Rate

The following table illustrates the results of the interviewing process. These represent a response rate of 50.4% and a refusal rate of 32.5% (calculated according to standard formulas developed by the Council of American Survey Research Organizations).

Disposition of all Telephone

Numbers Included in the Sample

Actual Percentage

Number  of Total 

Completed Interviews 1,001 14.3%

Refusals 452 6.5

Non-Working Number 3,117 44.6

No Answer (multiple attempts) 780 11.1

Business/Government. Number 341 4.9

Non-Eligible 841 12.0

Selected Respondent Unavailable 233 3.3

Deaf/Language Barrier 22 .3

Terminated During Interview 29 .4

Busy Signal (multiple attempts) 180 2.6

6,996 100.0%

 

Sample Characteristics

Furthermore, to accurately represent the population studied, it was necessary to constantly monitor the demographic composition (e.g., age, gender, household location) of the consumer sample throughout the data collection process. In actuality, the sample was composed of 47.6% men and 52.4% women. Interviews were conducted with persons ranging in age from 18 to 89.

The following table compares census population estimates with the sample distribution for key demographic characteristics.

The sample design and the quality control procedures used in the data collection ensure that the sample is representative. Thus, the findings may be generalized to the total population of service area consumers with high degree of confidence.

 

Error Rates Associated With Sample Sizes (95% Confidence Level)

ß Question Response Rate à

sample size ↓

10%

20%

30%

40%

50%

60%

70%

80%

90%

25

12.0

16.0

18.3

19.6

20.0

19.6

18.3

16.0

12.0

50

8.4

11.2

12.8

13.7

14.0

13.7

12.8

11.2

8.4

75

6.8

9.1

10.4

11.2

11.4

11.2

10.4

9.1

6.8

100

5.9

7.9

9.0

9.7

9.8

9.7

9.0

7.9

5.9

150

4.8

6.4

7.4

7.9

8.0

7.9

7.4

6.4

4.8

200

4.2

5.6

6.4

6.8

6.9

6.8

6.4

5.6

4.2

250

3.7

5.0

5.7

6.1

6.2

6.1

5.7

5.0

3.7

300

3.4

4.5

5.2

5.6

5.7

5.6

5.2

4.5

3.4

350

3.1

4.2

4.8

5.1

5.2

5.1

4.8

4.2

3.1

400

2.9

3.9

4.5

4.8

4.9

4.8

4.5

3.9

2.9

450

2.8

3.7

4.2

4.5

4.6

4.5

4.2

3.7

2.8

500

2.6

3.5

4.0

4.3

4.4

4.3

4.0

3.5

2.6

550

2.5

3.3

3.8

4.1

4.2

4.1

3.8

3.3

2.5

600

2.4

3.2

3.7

3.9

4.0

3.9

3.7

3.2

2.4

650

2.3

3.1

3.5

3.8

3.8

3.8

3.5

3.1

2.3

700

2.2

3.0

3.4

3.6

3.7

3.6

3.4

3.0

2.2

750

2.1

2.9

3.3

3.5

3.6

3.5

3.3

2.9

2.1

800

2.1

2.8

3.2

3.4

3.5

3.4

3.2

2.8

2.1

850

2.0

2.7

3.1

3.3

3.4

3.3

3.1

2.7

2.0

900

2.0

2.6

3.0

3.2

3.3

3.2

3.0

2.6

2.0

950

1.9

2.5

2.9

3.1

3.2

3.1

2.9

2.5

1.9

1000

1.9

2.5

2.8

3.0

3.1

3.0

2.8

2.5

1.9

1100

1.8

2.4

2.7

2.9

3.0

2.9

2.7

2.4

1.8

1200

1.7

2.3

2.6

2.8

2.8

2.8

2.6

2.3

1.7

1300

1.6

2.2

2.5

2.7

2.7

2.7

2.5

2.2

1.6

1400

1.6

2.1

2.4

2.6

2.6

2.6

2.4

2.1

1.6

1500

1.5

2.0

2.3

2.5

2.5

2.5

2.3

2.0

1.5

1600

1.5

2.0

2.2

2.4

2.5

2.4

2.2

2.0

1.5

1700

1.4

1.9

2.2

2.3

2.4

2.3

2.2

1.9

1.4

1800

1.4

1.8

2.1

2.3

2.3

2.3

2.1

1.8

1.4

1900

1.3

1.8

2.1

2.2

2.2

2.2

2.1

1.8

1.3

2000

1.3

1.8

2.0

2.1

2.2

2.1

2.0

1.8

1.3

 

 

 

2. Listing of Questionnaire Variables and Frequency Distribution

Frequency Distribution
United Way Region Crosstabulation
Region Crosstabulation

3. Bucks County Adult Household Survey 2001

 

 

 

4. Comparison of Adults Currently Smoking

 

5. Behavioral Risks
Return to the Table of Contents