CESS is committed to ensuring that all experiments we conduct meet the highest scientific and ethical standards. Accordingly, we have adopted a set of standardized guidelines and procedures for all of our centres. These procedures can be summarized briefly in the left tabs.
» CESS requires that the design of experiments conducted in its facilities, under its supervision, and with its subject pools, meet the highest scientific standards in experimental social science.
» For many seasoned experimenters a description of the experiment design in the application document (available at https://cess-nuffield.nuff.ox.ac.uk/experiments/for-researchers) will be sufficient. Nevertheless, these designs will be reviewed critically both by external reviewers and by the CESS staff. Poor design is a criterion for rejecting an experimental application.
» In the case of less seasoned experimenters or experimental novices, CESS strongly encourages these PIs to present their experimental designs in a CESS colloquium. The colloquium lasts one hour and is attended by members of the experimental research community at Oxford. It’s an invaluable opportunity to get very constructive feedback on an experimental design. These are held on a regular base during term time – requests for presenting at a CESS colloquium can be addressed to firstname.lastname@example.org.
» All experiments must meet CESS ethical standards – these are described at https://cess-web.nuff.ox.ac.uk/ethics/pages/guidelines.
» All experiments must be reviewed for ethical approval. Applications are submitted online at https://cess-web.nuff.ox.ac.uk/ethics/pages/home. Applications are reviewed by a minimum of two academic reviewers unaffiliated with Nuffield CESS. These reviews are typically completed within about two to three weeks. Ethical reviews are made available to PIs – PIs must address any shortcomings identified by reviewers before their experiments can proceed. Ethical reviews will be provided to appropriate third parties such as journal editors or IRBs.
» CESS strongly encourages all researchers conducting experiments at CESS facilities, or under CESS supervision, to submit pre-registration design and analysis plans as part of their application – this can be done at https://cess-web.nuff.ox.ac.uk/ethics/pages/program.
» Pre-registration plans are also submitted for external review. Pre-registration plans are reviewed by a minimum of two experimental social scientists unaffiliated with Nuffield CESS. These reviews are typically completed within about two to three weeks. Pre-registration plan reviews are made available to PIs. We expect all PIs to address any shortcomings reviewers identify in their pre-registration plans. Pre-registration reviews become part of the pre-registration record. They are time stamped as are the revisions adopted by the PIs. Pre-registration plans, and their external reviews, will be provided to appropriate third parties such as journal editors and funding agencies.
» All experimental data that is generated at CESS is subject to careful oversight by CESS managers – this includes lab, online, lab-in-the-field and field data. We guarantee the authenticity and integrity of any data that results from experiments conducted at CESS facilities.
» In the experimental lab our data oversight policy is the following:
- final programming of all experiments is conducted by CESS lab managers;
- results from experiments are initially compiled on CESS servers;
- these are then made available to PIs;
- these data plus proper documentation that includes the experimental protocols are archived at CESS;
- these data and protocols can be make available, in form that protects the identify of all subjects, to appropriate third parties that include the PIs of course but also Editors and funding agencies.
» Implementation of lab-in-the-field experiments by CESS follows the same protocol as for CESS conventional lab experiments. Data generation is subject to the same oversight and archiving policies.
» The data oversight policy for online experiments is very similar to the lab experiments:
- the data generation is subject to oversight by CESS online experiment managers;
- final programming of all online experiments is conducted by CESS managers on CESS servers;
- subject recruitment is managed by CESS;
- results from online experiments are initially compiled by CESS personnel on CESS servers;
- these data are then made available to PIs;
- these data plus proper documentation that includes experimental protocols and coding are archived at CESS;
- these data and protocols can be made available, in a form that protects the identify of all subjects, to appropriate third parties that include the PIs of course but also journal editors and funding agencies.
» CESS policy is to recommend to all PIs that they make their anonymized data, along with proper documentation, available to the public within 12 months, if not earlier.
» Maintaining data integrity in Randomized Control Trials/Field Experiments poses somewhat different challenges. Nevertheless CESS is committed to overseeing such data integrity for RCT/Field Experiments conducted under its auspices:
- RCTs/Field Experiments conducted under the auspices of CESS are also subject to strict CESS oversight. This includes of course the design phase but also the implementation of the field experiment and the compilation of resulting data.
- These final data are made available to PIs but also archived on CESS servers with appropriate documentation and are made available to appropriate third parties such as journal editors and funding agencies.