HS865 Unit 8 Convergent Mixed Methods Methodology Example

Survey and Interviews in Parallel Across Fourteen Sites: A Unit 8 Methodology Draft for a Study of Gonadal Shielding Habits

[Student Name]

Purdue University Global

HS865: Doctoral Project II: Research

Unit 8 Methodology Draft

[Instructor Name]

[Date]

Wrenfield Health and its figures are composite and shown as placeholders; no data have been collected, and the methods sources cited are real.

Research Questions

One: How often do radiographers in the network report placing a gonadal shield during anteroposterior pelvis radiography of children, and does this differ by site, years in practice and training program? Two: On which framework domains do routine shielders score differently from colleagues who have stopped? Three: How do radiographers describe the influences on their shielding decisions, and how do these accounts align with or differ from the survey results?

Design

The study uses a convergent mixed methods design, in which quantitative and qualitative data are collected in roughly the same period, analyzed separately and then merged to compare and interpret the results (Creswell & Plano Clark, 2018). A convergent design suits the question because the survey and interviews address the same determinants from different angles, and neither strand depends on the other's results to be collected. A sequential design was considered and rejected because it would add three to four months, and because the framework already tells the interview strand what to explore.

Setting

The Wrenfield Health imaging network includes fourteen sites performing projection radiography: two hospitals, four urgent care centers and eight outpatient imaging centers. All sites operate under the same radiology protocols, which were revised in [2021] to state that gonadal shielding is not routinely required. Each site has its own room layout and supply practices.

Participants

Survey: all registered radiographers employed by the network who perform projection radiography, about [286] people. No sampling; the survey is a census. Inclusion requires current registration and at least [one] shift a month in projection radiography. Students and radiographers working only in CT, MRI or mammography are excluded.

Interviews: [16] to [20] radiographers selected purposively from survey respondents who agree to be contacted, chosen to include people who report routine shielding and people who do not, across hospital and outpatient sites and across career stages. The number reflects guidance that most themes in a reasonably homogeneous sample appear within the first dozen interviews (Guest et al., 2006), with additional interviews allowing comparison between shielding groups.

Instruments

Survey. The survey has three parts. Part one asks about shielding practice for named examinations, beginning with the frequency item rewritten after the Unit 6 seminar, which names the pediatric AP pelvis examination and offers five responses from never to always. Part two measures the 14 framework domains using items adapted from a published questionnaire whose content validity was examined by assessing whether items discriminated between domains (Huijg et al., 2014), reworded for shielding and rated on a seven-point agreement scale. Part three records demographics and training, including the program where the respondent qualified and the year.

Adaptation of the questionnaire weakens the claim that the original validity evidence applies. To address this, [six] radiographers outside the network will complete cognitive interviews on the adapted items, and internal consistency will be reported for each domain in the study sample.

Interview guide. Semi-structured, with one or two prompts for each domain, ordered to begin with recent concrete cases, such as the last time the participant did or did not place a shield, before moving to beliefs and influences.

Each survey domain has a matching interview prompt, so that the merge can compare what radiographers rate with what they describe, domain by domain.

Procedures

The imaging education office, not the researcher, will send survey invitations by network email with two reminders over three weeks. The survey will take about [15] minutes. The final survey item invites respondents to volunteer for an interview through a separate form, so survey responses remain unlinked to names. Interviews will be conducted by video or in person in a private room, recorded with permission, transcribed and deidentified.

Analysis

Quantitative. Shielding frequency will be dichotomized as routine (often or always) versus not routine for the primary comparison. Domain scores will be compared between groups using Mann-Whitney tests, since domain scores from short scales are unlikely to be normally distributed, with effect sizes reported. Because fourteen separate tests make chance findings likely, the false discovery rate will be controlled with the Benjamini-Hochberg procedure (Benjamini & Hochberg, 1995). A logistic regression will then examine the domains that differ, adjusting for site type and years in practice.

Qualitative. Transcripts will be analyzed using directed content analysis, with the 14 domains as initial codes, a residual code for content outside the framework and a second coder for [25] percent of transcripts (Atkins et al., 2017).

Integration. Results will be merged in a joint display that lists each domain with its survey result and the interview themes coded to it, and labels each domain as convergent, complementary or discordant (Fetters et al., 2013).

Timeline

Cognitive interviews on the adapted items will take place in the first month after approval. The survey will be open for three weeks in the second month, with interviews scheduled during the same weeks and the following two, so that both strands are collected within about five weeks. Quantitative analysis and transcription will run in parallel in the third month, qualitative coding in the third and fourth, and integration in the fifth. Collecting both strands in one window is the main practical advantage of the convergent design: interview participants describe the same period of practice that the survey asks about, and no strand waits for the other.

Data Management

Survey data will be held on the university's secure platform without names or network identifiers. The interview contact form will be stored separately and destroyed after interviews are scheduled. Audio files will be erased once each transcript has been verified against them, and transcripts will carry participant codes only. Access to the coded files is limited to the researcher and the second coder.

Rigor

Quantitative rigor rests on the census approach, cognitive testing of adapted items and control of multiple comparisons. Qualitative rigor rests on the second coder, an audit trail of codebook changes and a reflexive journal. The researcher's role as a radiography educator in the region is disclosed, and no participant will be a former student of the researcher's program.

Limitations

The survey depends on self-report, which may understate a practice that participants know is discouraged. One network limits transfer. The convergent design means interview selection cannot be based on full survey results, only on the shielding item.

References

Atkins, L., Francis, J., Islam, R., O'Connor, D., Patey, A., Ivers, N., Foy, R., Duncan, E. M., Colquhoun, H., Grimshaw, J. M., Lawton, R., & Michie, S. (2017). A guide to using the Theoretical Domains Framework of behaviour change to investigate implementation problems. Implementation Science, 12, 77. https://doi.org/10.1186/s13012-017-0605-9

Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289-300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x

Creswell, J. W., & Plano Clark, V. L. (2018). Designing and conducting mixed methods research (3rd ed.). SAGE Publications.

Fetters, M. D., Curry, L. A., & Creswell, J. W. (2013). Achieving integration in mixed methods designs: Principles and practices. Health Services Research, 48(6 Pt 2), 2134-2156. https://doi.org/10.1111/1475-6773.12117

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18(1), 59-82. https://doi.org/10.1177/1525822X05279903

Huijg, J. M., Gebhardt, W. A., Crone, M. R., Dusseldorp, E., & Presseau, J. (2014). Discriminant content validity of a theoretical domains framework questionnaire for use in implementation research. Implementation Science, 9, 11. https://doi.org/10.1186/1748-5908-9-11