Reaching the U.S. Cell Phone Generation: Comparison of Cell Phone Survey Results with an Ongoing Landline Telephone Survey

Noncoverage rates in U.S. landline-based telephone samples dueto cell phone only households (i.e., households with no landlinebut accessible by cell phone) and the corresponding potentialfor bias in estimates from surveys that sample only from landlineframes are growing issues. Building on some of the few publishedstudies that focus on this problem, a study was conducted inthree states (Georgia, New Mexico, and Pennsylvania) as partof the Behavioral Risk Factor Surveillance System (BRFSS), theworld's largest ongoing public health telephone survey, to evaluatethe effectiveness of conducting the BRFSS interview with a sampledrawn from dedicated cell phone telephone exchanges and mixed-use(landline and cell phone) exchanges. Approximately 600 interviewswere conducted in each of two groups: cell phone only adults(n = 572) and adults with both a landline and a cell phone (n= 592). Making comparisons with data from the ongoing, landline-basedBRFSS survey, we report on response rates, demographic characteristicsof respondents, key survey estimates of health conditions andrisk behaviors, and survey costs. The methods used in this studyhave wide application for other U.S. telephone surveys.     

Assessment of Cigarette Smoking and Smoking Susceptibility among Youth: Telephone Computer-Assisted Self-Interviews versus Computer-Assisted Telephone Interviews

This experimental study assesses the effect of two survey methods,telephone audio computer-assisted self-interviewing (T-ACASI)and computer-assisted telephone interviewing (CATI), on self-reportsof smoking behavior and smoking susceptibility among adolescents12–17 years of age in California. In T-ACASI, participantslisten to prerecorded, computer-controlled questions and respondby pressing the keypad on a touch-tone telephone. In CATI, interviewersadminister the questions and enter responses into a computer.Prior research suggests that youth may be more likely to reportsensitive behaviors in a self-administered survey like T-ACASIcompared to an interviewer-administered survey like CATI, dueto greater perceived confidentiality. Logistic regression analyseswere conducted on unweighted data, controlling for demographicdifferences. Adjusted estimates of current smoking (past 30days) were significantly greater in T-ACASI (8.3 percent) thanCATI (4.5 percent). Smoking susceptibility (i.e., lack of afirm commitment not to smoke among those who have never smoked)was also greater in T-ACASI (45.0 percent) than CATI (34.9 percent).In both surveys, social desirability response bias was negativelyassociated with smoking, which suggests that response bias wasproblematic for both modes. Many respondents reported that aparent was present during the interview (59.4 percent in CATI;42.0 percent in T-ACASI). In both surveys, parental presencewas negatively associated with smoking, which suggests thatthis factor could also contribute to underreporting. Applicationof sample weights to the data eliminated the survey mode effects;however, the CATI current smoking estimate (9.3 percent) fromthis study was significantly less than an estimate (14.2 percent)obtained from a self-administered, school-based survey conductedthe same year on California adolescents.     

Evaluating the Effects of Screening for Telephone Service in Dual Frame RDD Surveys

The high costs and largely unknown error properties of cellulartelephone interviews make screening for cell-only adults a potentiallyattractive option in dual frame RDD surveys. Screening out adultswith landline telephones from the cellular sample does not affectthe coverage properties of a dual frame survey, but it may affectother sources of error, especially nonresponse. In this study,data from a 2006 dual frame RDD survey conducted for the PewResearch Center, the Associated Press, and AOL are used to evaluatethe effects of implementing a cell-only screener on both thebias and variance of weighted survey estimates. The effect ofscreening appears to be minimal so long as an adjustment fortelephone service is included in the weighting method. Resultsof an attempt to correct for residual nonresponse due to inaccessibilityare also discussed.     

Cannabis Use and Antisocial Behavior among Youth

Despite the numerous efforts to curb substance use and abuse through legislation and interventions, marijuana consumption continues to be a major social problem, particularly among young adults in the United States. We provide new information on the relationship between cannabis use and antisocial behavior by analyzing a sample of young adults (aged 18–20) from the National Epidemiological Survey of Alcohol and Related Conditions (NESARC). We examine a broad set of cannabis use patterns and multiple dimensions of antisocial behaviors and test the empirical importance of two prominent criminological theories—general strain and social bond—in explaining associations between cannabis use and antisocial behavior. We include important socioeconomic, demographic, health and health behaviors, and contextual information in all regressions to control for confounding factors. Our results imply that cannabis use is positively and significantly related to antisocial behavior among young adults, and general strain and social bond theories cannot fully explain our findings. As expected, the estimated association with antisocial behavior is stronger for more frequent cannabis users.     

Sources of health insurance and characteristics of the uninsured: analysis of the March 2007 Current Population Survey

This Issue Brief provides historic data through 2006 on the number and percentage of nonelderly individuals with and without health insurance. Based on EBRI estimates from the U.S. Census Bureau's March 2007 Current Population Survey (CPS), it reflects 2006 data. It also discusses trends in coverage for the 1994-2006 period and highlights characteristics that typically indicate whether an individual is insured. HEALTH COVERAGE CONTINUES DECLINE: The percentage of the nonelderly population (under age 65) with health insurance coverage continued to decline, reaching to a post-1994 low of 82.1 percent in 2006. Declines in health insurance coverage have been recorded in all but four years since 1994, when 36.5 million nonelderly individuals were uninsured; in 2006, the uninsured population was 46.5 million. EMPLOYMENT-BASED COVERAGE REMAINS DOMINANT SOURCE OF HEALTH COVERAGE: Employment-based health benefits remain by far the most common form of health coverage in the United States, consistently covering 60-70 percent of nonelderly individuals. In 2006, 62.2 percent of the nonelderly population had employment-based health benefits, as compared with 64.4 percent in 1994. Between 1994 and 2000, the percentage of the nonelderly population with employment-based coverage expanded. Since 2000, the percentage has declined. PUBLIC PROGRAM COVERAGE IS STABLE: Public-sector health coverage was slightly lower as a percentage of the population in 2006, accounting for 17.5 percent of the nonelderly population. The decline was due to a drop in the percentage of the population covered by the Tricare/CHAMPVA program. Enrollment in Medicaid and the State Children's Health Insurance Program increased, reaching 34.9 million in 2006, and covering 13.4 percent of the nonelderly population, which is significantly above the 10.5 percent level of 1999, but not far above the 12.7 percent level of 1994. INDIVIDUAL COVERAGE STABLE: Individually purchased health coverage was unchanged in 2006 and has basically hovered in the high 6 and low 7 percent range since 1994. PRIVATE- VS. PUBLIC-COVERAGE TRENDS REVERSING: Health insurance coverage generally has not sustained unbroken trends since 1994. There were crosscurrents: Employment-based coverage expanded significantly in the 1994-2000 period to exceed the growth in public programs. Subsequently, the dynamic reversed, as public programs expanded while employment-based coverage declined. It appears that 2005 might be the beginning of a new trend, where the erosion in employment-based coverage is not being offset by expansions in public programs. This may be due to the fact that, while unemployment is relatively low, the cost of providing health benefits continues to increase faster than inflation.     

BIAS IN LIST-ASSISTED TELEPHONE SAMPLES

A number of researchers have suggested list-assisted samplingfor the selection of telephone households to overcome some ofthe operational difficulties associated with the Mitofsky-Waksbergmethods of random digit dialing (RDD). An advantage of a list-assistedmethod of RDD is that an equal probability systematic sampleof telephone numbers can be selected and the variances of estimatesfrom such a sample are usually lower than from a clustered designlike the Mitofsky-Waksberg method. The main disadvantage ofthe list-assisted method is that it excludes some householdsfrom the sample, thus creating a coverage bias in the estimates.This article describes research on the coverage bias for a particularmethod of list-assisted sampling. The two key determinants ofcoverage bias are the proportion of households that are noteligible for the sample and the differences in the characteristicsof the covered and not covered populations. The results showthat about 4 percent of all households are excluded in nationalsamples using this method of sampling. Furthermore, they showthat the differences between the covered and uncovered populationsare generally not large. The coverage bias resulting from theseconditions may often be small.     

Risk Factors for Problem Gambling in California: Demographics,Comorbidities and Gambling Participation

While population surveys have been carried out in numerous jurisdictions internationally, little has been done to assess the relative strength of different risk factors that may contribute to the development of problem gambling. This is an important preparatory step for future research on the etiology of problem gambling. Using data from the 2006 California Problem Gambling Prevalence Survey, a telephone survey of adult California residents that used the NODS to assess respondents for gambling problems, binary logistic regression analysis was used to identify demographic characteristics, health-related behaviors, and gambling participation variables that statistically predicted the odds of being a problem or pathological gambler. In a separate approach, linear regression analysis was used to assess the impact of changes in these variables on the severity of the disorder. In both of the final models, the greatest statistical predictor of problem gambling status was past year Internet gambling. Furthermore, the unique finding of a significant interaction between physical or mental disability, Internet gambling, and problem gambling highlights the importance of exploring the interactions between different forms of gambling, the experience of mental and physical health issues, and the development of problem gambling using a longitudinal lens.     

Consequences of reducing nonresponse in a national telephone survey

Critics of public opinion polls often claim that methodological shortcuts taken to collect timely data produce biased results. This study compares two random digit dial national telephone surveys that used identical questionnaires but very different levels of effort: a "Standard" survey conducted over a 5-day period that used a sample of adults who were home when the interviewer called, and a "Rigorous" survey conducted over an 8-week period that used random selection from among all adult household members. Response rates, computed according to AAPOR guidelines, were 60.6 percent for the Rigorous and 36.0 percent for the Standard study. Nonetheless, the two surveys produced similar results. Across 91 comparisons, no difference exceeded 9 percentage points, and the average difference was about 2 percentage points. Most of the statistically significant differences were among demographic items. Very few significant differences were found on attention to media and engagement in politics, social trust and connectedness, and most social and political attitudes, including even those toward surveys.     

Bias in a Directory Sample for a Mail Survey of Rural Households

This study examines bias in a mail survey of rural householdsthat used telephone directories as the sampling frame and comparesit with a parallel survey that used an independent sample selectedfrom field listings. A higher response rate was obtained withthe directory sample, which was primarily attributed to pooreraddress information for the listed sample. Although coverageof the study population was higher for the listed frame thanfor the directory frame, no significant differences were foundfor eight respondent and seven household characteristics. Also,no significant differences were found for seven key contentvariables regarding health behaviors and attitudes. The findingssupport the use of telephone directories as the sampling framefor general population mail surveys of rural areas where suchcoverage of the population is considered adequate.     

Designing a household survey to address seasonality in child care arrangements

In household telephone surveys, a long field period may be required to maximize the response rate and achieve adequate sample sizes. However, long field periods can be problematic when measures of seasonally affected behavior are sought. Surveys of child care use are one example because child care arrangements vary by season. Options include varying the questions posed about school-year and summer arrangements or posing retrospective questions about child care use for the school year only. This article evaluates the bias associated with the use of retrospective questions about school-year child care arrangements in the 1999 National Survey of America's Families. The authors find little evidence of bias and hence recommend that future surveys use the retrospective approach.     

Effects and Costs of Tracing Strategies on Nonresponse Bias in a Survey of Workers with Low-Back Injury

Declining rates of participation are an increasing challengefor studies that involve telephone surveys. This study examinedthe costs of a telephone survey methodology that used increasinglyintensive tracing methods to track a pool of claimants who hadsustained occupational back injuries. It also compared the respondentsample to people who refused the survey and/or were not locatedor contacted. 3,181 claimants were drawn from a database maintainedby the Missouri Division of Workers’ Compensation (DWC)and 1,475 completed a telephone interview. The DWC databaseprovided data reflecting monetary and disability outcomes forall potential participants; telephone interviews provided additionalself-reported data. More intensive tracing strategies improvedthe representativeness of the sample, yielding more women andminority participants. Relative to less intensive techniques,advanced tracing efforts located people whose DWC records weremore similar to claimants not located. While the hands-on tracingefforts reduced the apparent bias of the respondent sample,costs increased as tracing strategies intensified – over$98,000 was spent to trace 1,027 claimants who were never locatedor contacted. The results suggest some guidelines that may facilitatedecision-making for researchers and funders who must balancethe trade-off between costs and nonresponse bias when planningsurveys.     

What's Missing from National Landline RDD Surveys?: The Impact of the Growing Cell-Only Population

The number of cell phone only households has continued to grow– 12.8 percent of all households by the end of 2006, accordingto the National Health Interview Survey (NHIS). To assess theextent and nature of the potential bias in landline telephonesamples created by the absence of cell-only households, thePew Research Center conducted four independent dual frame studiesin 2006, three of which included at least 200 interviews withcell-only respondents. Despite the fact that there are significantdifferences between cell-only and landline respondents on manyimportant variables, across the four surveys with differentsubstantive content we find that including a cell-only samplewith a landline RDD sample produces general population estimatesthat are nearly identical to those from the landline samplealone. Yet, while the noncoverage problem is currently not damagingestimates for the entire population, we find evidence that itdoes create biased estimates on certain variables for youngadults, 25 percent of whom are cell-only according to the mostrecent government estimate.     

Does Telephone Audio Computer-Assisted Self-Interviewing Improve the Accuracy of Prevalence Estimates of Youth Smoking? Evidence from the UMass Tobacco Study

Despite their advantage for obtaining representative samplesof adolescents, telephone surveys have been regarded as an inferiormethod for collecting data on youth tobacco use because theyyield lower estimates than school-based, self-administered surveys.Although no gold standard for smoking estimates exists, thelower estimates in telephone surveys have been attributed tounderreporting due to youths’ concern that parents orothers may overhear their responses. Telephone audio computer-assistedself-interviewing (T-ACASI) is a cost-effective method for obtaininga representative sample of youths and provides increased privacyfor the respondent. We hypothesize that using T-ACASI wouldencourage youths to report more fully smoking behavior comparedto traditional interviewer-administered telephone methods. Ouranalysis further assesses whether respondent age, gender, race/ethnicity,and parental attitude toward smoking moderated the relationshipbetween survey mode and smoking reports. Using data from a statewidetobacco use survey that randomly assigned youth respondentsto either T-ACASI or interviewer-administered modes, we findthat youths were more likely to report smoking behaviors inT-ACASI mode and that this was especially true for girls, particularlythose who believed their parents would disapprove strongly oftheir smoking. The findings suggest that traditional telephonesurveys may underestimate smoking prevalence in most girls bya factor of two, and that a technique for ensuring privacy forthese respondents is an important component of effective telephonesurvey methodology.     

THE ANSWERING MACHINE GENERATION: WHO ARE THEY AND WHAT PROBLEM DO THEY POSE FOR SURVEY RESEARCH?

Increased incidence of telephone answering machines and theuse of such devices to screen calls pose a potential threatto the representativeness of samples in telephone surveys. Usingdata from nine statewide surveys, this analysis examines theextent to which answering machines are used to screen callsand the demographic characteristics associated with answeringmachine use and call screening. Results show that at most twoto three percent of households use answering machines to screencalls, and that such screening is more likely to take placein households with higher family incomes, outside rural areas,and which include individuals who are younger and have higherlevels of education. While call screening does not presentlythreaten the representativeness of samples in telephone surveys,the increased incidence of answering machines together withthe increased percentage of households indicating these devicesare sometimes used to screen calls demonstrate that the potentialbias from this source is growing.     

Low-Income Children's Participation in a Public Health Insurance Program in Georgia

To expand health care coverage to uninsured, low-income children, the Congress created the State Children's Health Insurance Program (SCHIP) in 1997. Given ample evidence that state Medicaid programs have failed to enroll many eligible children, experts questioned whether SCHIP could successfully enroll low-income children. Using Georgia state SCHIP data, we analyzed enrollment patterns and identified factors contributing to program enrollment. This study found that the children's enrollment is explained by perceived costs and benefits of joining the program as well as the underlying family structure, demographic factors, and health status. We discuss implications of these findings on the participation of low-income families in public sector programs.     

Expanding health insurance for children: examining the alternatives

This Issue Brief examines the issue of uninsured children. The budget reconciliation legislation currently under congressional consideration earmarks $16 billion for new initiatives to provide health insurance coverage to approximately 5 million of the 10 million uninsured children during the next five years. Proposals to expand coverage among children include the use of tax credits, subsidies, vouchers, Medicaid program expansion, and expansion of state programs. However, these proposals do not address the decline in employment-based health insurance coverage--the underlying cause of the lack of coverage, to the extent that a cause can be identified. What is worse, some proposals to expand health insurance among children may discourage employers from offering coverage. Between 1987 and 1995, the percentage of children with employment-based health insurance declined from 66.7 percent to 58.6 percent. Despite this trend, the percentage of children without any form of health insurance coverage barely increased. In 1987, 13.1 percent were uninsured, compared with 13.8 percent in 1995. Medicaid program expansions helped to alleviate the effects of the decline in employment-based health insurance coverage among children and the potential increase in the number of uninsured children. Between 1987 and 1995, the percentage of children enrolled in the Medicaid program increased from 15.5 percent to 23.2 percent. Some questions to consider in assessing approaches to improving children's health insurance coverage include the following: If the government intervenes, should it do so through a compulsory mechanism or a voluntary system? Is the employment-based system "worth saving" for children? In other words, are the market interventions necessary to keep this system functioning for children too regulatory, too intrusive, and too cumbersome to be practical? In addition to reforming the employment-based system, what reforms are necessary in order to reach those families who have no coverage through the work place? Which approaches are both efficient and politically acceptable? Employment-based coverage of children will likely continue. The challenge for lawmakers is to find a way to cover more uninsured children without eroding employment-based coverage. Several current legislative proposals attempt to avoid this problem by excluding children who have access to employment-based coverage. Without such a requirement, the opportunity to purchase coverage at a discount would create incentives for some low-income employees to drop dependent/family coverage, which in turn could lead some employers to drop their health plans.     

ESTIMATING TELEPHONE NONCOVERAGE BIAS WITH A TELEPHONE SURVEY

Nontelephone households are implicitly treated as a static populationin discussions of sampling frame noncoverage. Yet telephoneservice is known to be episodic for many households, who maygain or lose service as their financial situation changes orwhen they move. Thus the population of telephone householdsat any given time includes households that were recently a partof the nontelephone population. These households may be usedto characterize the nature of some noncoverage errors and evento estimate their magnitude. Using a panel constructed withthe 1992-93 Current Population Survey, "transient" telephonehouseholds-those who gained or lost service over the year coveredby the panel-are shown to comprise over half of the panel householdsreporting no telephone service in either the 1992 or 1993 surveys.These households are compared with the total nonphone populationand found to be similar on a variety of key demographic characteristics.Several statewide Virginia telephone surveys are used to comparehouseholds reporting "intermittent" phone service with nontelephonehouseholds surveyed through in-person interviews. Householdsreporting intermittent telephone service were very similar tonontelephone households in terms of health insurance coverageand other variables known to be related to telephone status.     

SAME-GENDER SEX IN THE UNITED STATES IMPACT OF T-ACASI ON PREVALENCE ESTIMATES

Well-conducted telephone surveys provide an economical meansof estimating the prevalence of sexual and reproductive behaviorsin a population. There is, however, a nontrivial potential forbias since respondents must report sensitive information toa human interviewer. The National STD and Behavior MeasurementExperiment (NSBME) evaluates a new survey technology—telephoneaudio computer-assisted self-interviewing (T-ACASI)—thateliminates this requirement. The NSBME embedded a randomizedexperiment in a survey of probability samples of 1,543 U.S.and 744 Baltimore adults ages 18 to 45. Compared with NSBMErespondents interviewed by human interviewers, respondents interviewedby T-ACASI were 1.5 to 1.6 times more likely to report same-gendersexual attraction, experience, and genital contact. The impactof T-ACASI was more pronounced (odds ratio = 2.5) for residentsof locales that have historically been less tolerant of same-gendersexual behaviors and for respondents in households with children(odds ratio = 3.0).     

Moving beyond poverty: neighborhood structure, social processes, and health

We investigate the impact of neighborhood structural characteristics, social organization, and culture on self-rated health in a large, cross-sectional sample of urban adults. Findings indicate that neighborhood affluence is a more powerful predictor of health status than poverty, above and beyond individual demographic background, socioeconomic status, health behaviors, and insurance coverage. Moreover, neighborhood affluence and residential stability interact in their association with health. When the prevalence of affluence is low, residential stability is negatively associated with health. Neighborhood affluence also accounts for a substantial proportion of the racial gap in health status. Finally, collective efficacy is a significant positive predictor of health but does not mediate the effects of structural factors.