Quantitative, Qualitative, Comparative, and Historical Methodologies Commons^™

How Customization Affects Survey Interaction, Antje Rosebrock, Malte Schierholz

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

One common trend in the world of survey data collection is the increasing use of new technological developments which can change the nature of the survey interview. A fairly recent trend is the use of machine-learning techniques to customize questions for respondents. This has the potential to create an individualized experience for the respondent and to improve data quality. Nevertheless, little is known so far of how customization affects the interaction in the survey interview.

We introduce a tool developed by Schierholz et al. (2018) to code respondents’ occupation categories during the survey. The tool uses supervised learning algorithms to …

Can Paradata Predict Interviewer Effects?, Sharan Sharma

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Consideration of interviewer effects (interviewer measurement error variance) in active quality control does not seem widespread despite its known effect on reducing precision of survey estimates. One major obstacle is that interviewer effect estimates computed on partial data (as a survey is in progress) can be very unstable. We address this issue by exploring the use of paradata (keystrokes and time stamps generated during the computer-assisted interviewing process) as proxies of interviewer effects with a focus on large-scale repeated cross-section or panel surveys.

We first estimate interviewer effects for each item in our analysis by using multilevel models that include …

Humans Vs. Machines: Comparing Coding Of Interviewer Question-Asking Behaviors Using Recurrent Neural Networks To Human Coders, Jerry Timbrook, Adam Eck

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Standardized survey interviewing techniques are intended to reduce interviewers’ effects on survey data. A common method to assess whether or not interviewers read survey questions exactly as worded is behavior coding. However, manually behavior coding an entire survey is expensive and time-consuming. Machine learning techniques such as Recurrent Neural Networks (RNNs) may offer a way to partially automate this process, saving time and money. RNNs learn to categorize sequential data (e.g., conversational speech) based on patterns learned from previously categorized examples. Yet the feasibility of an automated RNN-based behavior coding approach and how accurately this approach codes behaviors compared to …

Analysing The Influence Of Non-Observable And Observable Interviewer Characteristics On Measurement Error: Evidence From Zambia, P. Linh Nguyen

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

In Sub-Saharan Africa, where only one in five people uses the Internet and connectivity issues restrict the possibility for phone surveys in rural areas, interviewer-administered face-to-face (F2F) surveys are and will remain the principal data collection tool in the foreseeable future. Yet questions remain as to what extent previous findings on interviewer-administered surveys from Western countries may apply to a different cultural and geographical context. In this light, the objective of this study is to investigate the influence of certain observable interviewer characteristics (such as gender, age) and non-observable characteristics (such as education, attitudes) on interviewer variance on a subset …

Examining Interviewer Effects On The Agricultural Labor Survey: A Mixed-Methods Approach, David Biagas, Emilola Abayomi, Joseph Rodhouse, Heather Ridolfo

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewer effects remain a pervasive problem in survey research. Utilizing a mixed-methods approach, this study explores the effects that interviewers have on the reporting of agricultural workers on the USDA’s Agricultural Labor Survey. The Agricultural Labor Survey is administered by the USDA’s National Agricultural Statistics Service on a biannual basis, with each data collection period collecting information for two quarters (e.g. April and January of 2018). While the majority of data is collected via computer-assisted telephone interviewing, a sizeable proportion is completed online or via mail. Using quantitative and qualitative methods, this study explores the patterns of bias introduced by …

Did The Respondent Really Mean That? How The Behaviors Of Cati Interviewers And Data Editors Impact Measurement And Processing Errors In Establishment Surveys, Joseph Rodhouse, Heather Ridolfo, Emilola Abayomi, David Biagas

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

It is well documented that interviewers can have profound effects on the survey data collection process. This research looks to build on that knowledge by examining the relationship between CATI interviewers and data editors and how the recording of answers, and editing of the survey answers recorded, contribute to total survey error (TSE). Specifically, we are interested in comparing answers recorded by CATI interviewers and the final response codes after the editing stage. Since the interviewing stage and the editing (or processing) stage are often two distinct phases of the data collection process where the interviewer and data editor work …

How To Conduct Effective Interviewer Training: A Meta-Analysis, Jessica Daikeler

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewer training can improve the performance of interviewers and thus also the quality of survey data. However, the question of how effective interviewer training is for improving data quality and more importantly, which determinates drive its success, remain unanswered. This research uses meta-analytical methods to evaluate both the improvements in data quality due to interviewer training and the effectivity of training modules with respect interviewer performance. We consider various aspects of data quality, namely unit nonresponse, item nonresponse, probing behavior, administration, reading, and recording. Based on more than sixty experimental studies, we find that comprehensive interviewer training improves unit- and …

Response Times As An Indicator Of Data Quality: Associations With Interviewer, Respondent, And Question Characteristics In A Health Survey Of Diverse Respondents, Dana Garbarski, Jennifer Dykema, Nora Cate Schaeffer, Dorothy Farrar Edwards

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Survey research remains one of the most important ways that researchers learn about key features of populations. Data obtained in the survey interview are a collaborative achievement accomplished through the interplay of the interviewer, respondent, and survey instrument, yet our field is still in the process of comprehensively documenting and examining whether, when, and how characteristics of interviewers, respondents, and questions combine to influence the quality of the data obtained.

Researchers tend to consider longer response times as indicators of potential problems as they indicate longer processing or interaction from the respondent, the interviewer (where applicable), or both. Previous work …

Exploring The Impact Of Interviewer Perceptions And Interviewer-Respondent Interactions On The Survey Of Income And Program Participation: Analysis Of Cari Recordings, Erica Yu, Rodney L. Terry, Alina Kline, Holly Fee, Robin Kaplan

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewers play a significant role in telephone and face-to-face interviews, including gaining respondent cooperation and administering survey questions. Increasingly, interviewers’ perceptions of the respondent and interview experience, such as cooperativeness and interest, are also being used to assess measurement error and make adjustments to data (West, 2013; Kirchner et al., 2017). Although interviewer perceptions are typically recorded at the end of the interview, interviewers are likely to begin forming perceptions about the household and respondent based on their first contact attempt (and continue developing them during the interview). We hypothesize that interview context factors, such as interviewer perceptions of the …

What Do Interviewers Learn? Changes In Interview Length And Interviewer Behaviors Over The Field Period, Kristen Olson, Jolene Smyth

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewers systematically speed up over the field period of a survey as they conduct interviews (Olson and Peytchev 2007; Olson and Bilgen 2011; Kirchner and Olson 2017). Competing hypotheses for this increase in speed is that interviewers learn from previous interviews, changing their behaviors accordingly, or that they change behaviors in response to who the respondent is, including both respondent’s fixed characteristics and their response propensity. Previous work (e.g., Kirchner and Olson 2017) has failed to completely explain this learning effect, even after accounting for a wide range of measures of each of these hypotheses. However, prior work has not …

Scientific Network Of Experts: Interviewer Effects And Interviewer Training, Daniela Ackermann-Piek, Joe Sakshaug

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Although the collection of survey data is undergoing a notable shift toward online and mixed-mode data collection methods (Baker et al., 2010; Groves, 2011), interviewers are still heavily involved in the majority of survey data collections that serve as a basis for important economic, educational, and public policy decisions. Research supports the notion that interviewer characteristics and task-specific skill levels significantly influence the resulting data quality (see, e.g. Ackermann-Piek, 2018; Billiet & Loosveldt, 1988; Dahlhamer, Cynamon, Gentleman, Piani, & Weiler, 2010; Durand, 2005; Fowler Jr., 1991; Hox & de Leeuw, 2002; Jäckle, Lynn, Sinibaldi, & Tipping, 2013; Sakshaug, Tutz, & …

Investigating The Use Of Nurse Paradata In Understanding Nonresponse To Biological Data Collection, Fiona Pashazadeh, Alexandru Cernat, Joseph W. Sakshaug

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

The recent collection of biological data in large-scale sample surveys has opened up new possibilities for research into the interactions between physical and social mechanisms in the general population. Whilst the possibilities are undoubtedly exciting, these data can create additional challenges from the viewpoints of both collection and analysis. In particular, the extra burden of biological data collection can lead to increased incidences of nonresponse, potentially affecting the quality of the data and the robustness of results from subsequent analysis. Where the two-stage nurse visit survey design is used, such as in Understanding Society (UKHLS) and the English Longitudinal Study …

Nvestigating The Utility Of Interviewer Observations On The Survey Response Process, Frauke Kreuter, Brady West, Ting Yan, Michael Josten, Heather Schroeder

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

TBD.

Modeling Interviewer Effects In A Large National Health Study, James Dahlhamer, Aaron Maitland, Benjamin Zablotsky, Carla Zelaya

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewers play a critical role in determining the quality of data collected in face-to-face surveys. Interviewers can have positive effects on recruiting sample members to participate, leading to higher response rates. Conversely, interviewers can have negative effects on the quality of measurement. The literature suggests that interviewers can bias answers when observable characteristics of the interviewer influence the respondent to answer questions a certain way. For example, the sex or race of interviewers may influence respondents’ answers about their own attitudes toward sex or race. However, it is more common for differences in interviewer behavior, such as how questions are …

Questions Administered By Telephone Or In Person: Differences In Interviewer-Respondent Interactions, Yfke Ongena, Marieke Haan

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

When choosing a mode for data collection of computer-assisted surveys, a researcher has three main options available: the computer-assisted telephone interview (CATI), the computer-assisted personal interview (CAPI) or a web interview (i.e., a self-adminstered interview). Generally, CAPI allows for collecting most complex data, of the highest quality, but only when interviewers are well-trained and effort is made to monitor and manage interviewers during field work. This higher data quality in CAPI interviews may be due to the finding that presence of an interviewer reduces the amount of respondents’ satisficing behaviors (i.e., not investing the required effort to provide meaningful answers, …

Interviewer Effects On Data Quality: Does Interviewer Experience Favor Completions Over Quality?, Megan M. Ruxton, Rodney Muilenburg

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Interviewer experience has been identified as an important factor in achieving higher response rates for telephone interviews. The causal mechanisms underlying this relationship remain unclear (Couper and Groves 1992; Jackle, Lynn, Sinibaldi and Tipping 2013; West and Blom 2016), but extant research suggests a combination of experience and personality traits, skills and attitudes explain substantial variation in cooperation rates (Groves and Couper; Jackle et al 2013). West and Blom (2016) summarize the positive relationship between experience and response rates and call for work to identify additional mediators of these relationships. The relationship between experience and data quality is even less …

Modelling Group-Specific Interviewer Effects On Nonresponse Using Separate Coding For Random Slopes In Multilevel Models, Jessica M. E. Herzing, Annelies G. Blom, Bart Meuleman

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

To enhance response among underrepresented groups and hence, to increase response rates and to decrease potential nonresponse bias survey practitioners often use interviewers in population surveys (Heerwegh, 2009). While interviewers tend to increase overall response rates in surveys (see Heerwegh, 2009), research on the determinants of nonresponse have also identified human interviewers as one reason for variations in response rates (see for examples Couper & Groves, 1992; Durrant, Groves, Staetsky, & Steele, 2010; Durrant & Steele, 2009; Hox & de Leeuw, 2002; Loosveldt & Beullens, 2014; West & Blom, 2016). In addition, research on interviewer effects indicates that interviewers introduce …

Do You Need A Foot-In-The-Door Or Is A Toe Enough? Scripting Introductions That Induce Tailoring And Increase Participation In Telephone Interviews, Kim Ethridge, Matt Jans, Matthew D. Mcdonough, Sam Vincent, Jamie Dayton, Naomi Freedner, Randal Zuwallack, Josh Duell, Don Allen, Dan Bertuna, Lew Berman, Mark Serafin, Kristin Reichl, Anneke Jansen, Wendi Gilreath

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Substantial research and practical experience shows that a telephone interviewer is most successful at gaining cooperation and avoiding refusals when they are free to tailor their introductory pitch to the potential respondent or household informant they reach. However, survey designers are often uncomfortable allowing interviewers to work “off-script,” and instruct interviewers to read introductory text verbatim. Further, some interviewers report being more comfortable with a script than without one. To bridge this gap between research and practice we asked, “Can we create a scripted introduction that engages the potential respondent, gets a foot-in-the-door, and facilitates interviewer tailoring?” This paper reports …

Unintended Interviewer Bias In A Community-Based Participatory Research Randomized Control Trial Among American Indian Youth, Patrick Habecker, Jerreed Ivanich

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

Community-based participatory research (CBPR) projects often employ members of the host partner community to engage and assist with research projects. However, CBPR may also introduce bias to survey statistics when community partners work as interviewers for projects within their own communities. Here, the advantage of employing interviewers from the local community and region may lead to unintended bias when participants and interviewers know each other outside of the research project. In situations where a preexisting social relationship exists, there is a greater possibility of social desirability bias. This may be particularly true for sensitive issues where they may not wish …

Implementing A Case Ownership Model With Telephone Interviewers, Jamie Wescott

Interviewer Workshop, 2019: Interviewers and Their Effects from a Total Survey Error Perspective

In an interviewer-administered survey, interviewers themselves can have a substantial impact on survey quality. In their review of the literature, West and Blom (2017) described the innumerable articles dedicated to describing this effect. A subset of these have provided support for the linkage between interviewer variation, such as differences in experience level and contacting approach, and variation in unit noncontact and nonresponse rates; these include Purdon, Campanelli, and Sturgis (1999), Groves and Couper (1998), and Blom (2012). In telephone studies, automated call scheduling systems help to mitigate these effects by using algorithms to ensure that cases receive an appropriate number …