SRO ensures high-quality data delivery through rigorous checks, coding, and privacy protection measures, using advanced software tools.

SRO ensures delivery of high quality data files using a combination of checks built into the questionnaire instruments, post-collection review of internally inconsistent responses, and a check-coding and data processing operation. Data processing steps include reconciliation of differences in data structure between modes of data collection.  SRO provides data cleaning, labeling of variables, and formatting of data for public release, including calculation of summary variables for analysis as needed.

Data Entry

SRO performs data entry for studies using paper questionnaires and also studies that incorporate other paper forms (such as time diaries, consent forms, etc.). While SRO does not currently have in-house optical scanning capability, SRO works with external vendors to define scanning criteria and provide quality assurance on returned scans.

Coding

SRO will provide coding for open-ended questions. An experienced staff of coders and data entry staff design and build code frames, implement coding, and oversee quality control checks. Production coding and data entry are done using the SRO Coding App, which facilitates working with interview data and performs a series of automatic checks designed to eliminate wild or inconsistent codes. SRO coding of open-ended text responses also includes review of responses and redaction of Personally Identifiable Information (PII) required to reduce the risk of respondent identification.

In addition, SRO employs several Computer-Assisted Qualitative Data Analysis Software tools, including NVivo and WordStat, to conduct text analytics including generating dictionary- and machine learning-based coding.

Data Processing

Datasets can be delivered in a range of formats as required for analysis. Raw data identifying respondents, including geographic information, are separated from survey data to protect the confidentiality of the respondent. Other case-level process data, paradata or interviewer characteristics may be appended to survey data files when explicitly approved by the University of Michigan IRB, and other relevant review boards. SRO adheres to the University of Michigan Institute for Social Research Policy on Safeguarding Respondent Privacy for all data delivery, including, but not limited to secure data transmission, separation of PII, and signed acceptance from the principal investigator to maintain and protect respondent confidentiality.