Skip to content

Instantly share code, notes, and snippets.

@rpietro
Created October 11, 2013 22:52
Show Gist options
  • Save rpietro/6943242 to your computer and use it in GitHub Desktop.
Save rpietro/6943242 to your computer and use it in GitHub Desktop.
Methods manuscript template for secondary data analysis

Methods for secondary data analyses

The Methods template presents a common structure used as a way to engage readers of peer-reviewed articles. Of importance, the structure to be presented here is one among many structure that might work, but this is certainly not the only one.

The length of each section in this template is not stipulated, primarily because length will depend on the target journal and what might have been established as socially acceptable within that local scientific community. This means that all four text blocks presented below could be written in a single paragraph or in two pages.

This template contains multiple elements from the STROBE and SAMPL reporting standards.

Design

  • longitudinal or cross sectional
  • prospective or retrospective
  • previous studies
  • part or the whole section can be in the last section of the Introduction, and where you place it is less important than actually including it

Examples

Ethics

  • Approval by Institutional Review Board
  • Protected Health Information and 18 elements

Examples

Database description

  • Sampling and target population
  • How data were collected and organization in charge
  • Data access
  • Validation studies

Examples

Variables

Outcome variables

  • whatever you might be measuring

Examples

Predicting and intervention variables

  • observed risk factors
  • interventions such as treatments

Examples

Potential confounders and interactions

  • Descriptive variables that are not potential confounders could be included in a separate sections

Examples

Strata

  • Watch out for total number of tables and graphics, easy to overshoot

Examples

Data analysis

  • Whenever possible, use graphics, with one main graphic corresponding to each main objective

Descriptive analysis

  • Describe your sample so that a peer or somebody will put your results to practice can evaluate whether your results actually match their own population

Examples

Modeling

  • Adjusted analyses to control for confounding and reduce bias
  • Details about whatever models might be used: propensity scores, generalized linear models, survival models

Examples

Reproducible research standards

  • New but extremely important trend

Examples

References

  1. STROBE
  2. SAMPL

licensed under a Creative Commons

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment