End of this page section.

Begin of page section: Contents:

Open Science

The Doctoral School of the Department of Psychology and the Graz Open Science Initiative (GOSI, https://psychologie.uni-graz.at/en/social-psychology/gosi/) encourage doctoral students to apply transparency and Open Scientific (OS) practices in their research.

Here, you can find information and materials for the implementation of open scientific practices. These are in line with the Transparency and Openness Promotion (TOP) Guidelines (see https://topfactor.org/).

authored by
the Head of the Doctoral School,
the Doctoral Student Representatives
and GOSI, 2023

Ten years ago, Psychology and the Social Sciences were shaken through the advent of the Replication Crisis: Methodologists (e.g., Simmons et al., 2011) suspected that a lot of research in the field is not reproducible because researchers and journals tend to prefer positive (“significant”) results for a publication. Researchers may even adjust their theory and hypotheses after they have looked into the data (known as “hypothesizing after the results are known” or “HARKing”) or run multiple unreported tests to find a positive result (known as “p-hacking”) for their studies in order to publish a “clean story” (see this Figure: https://onlinelibrary.wiley.com/cms/asset/b1fd8ca4-5445-4ec7-b9a0-68908becae55/brv12315-fig-0003-m.jpg).

This suspicion was confirmed through the “Replication Project: Psychology”, which showed that from a sample of almost 100 published studies with positive results only 39 replicated, when independent researchers did the studies again.

To tackle this problem, researchers called for more transparency. In this regard, the application of OS practices is important and useful, as they promote the reproducibility and replicability of analyses and studies, respectively, to counteract publication bias (e.g., Scheel et al., 2021, Banks et al., 2019).

 

There are four important OS practices to achieve more transparency of your research:

1. Preregistration of studies before you analyze your data; this will make the research process – from theory, to hypotheses, to analysis, to the results – more transparent. You can also submit a preregistration protocol to journals to receive a peer review - this is called a Registered Report.

2. Materials can be made openly accessible (or at least findable, accessible, interoperable, and reusable - FAIR) to ensure that other researchers understand how you set up your study.

3. Data can be made openly accessible (or at least findable, accessible, interoperable, and reusable – FAIR) after it has been anonymized. Along with the analysis code, this promotes reproducibility of one’s research.

4. Open Access of the final research article implies that it can be accessed and by read by everyone after its publication.

Before the data are collected for a study (or before the researchers look into the data for the first time), researchers can do a preregistration, which is a written protocol of their intended study.

In this protocol, they should indicate their confirmatory hypotheses (or indicate that the research is purely exploratory), variables, anticipated sample size (justified through, e.g., power analysis) and analysis plans.

Preregistration is not intended to bind researchers to a single preregistered analysis. Instead, a preregistration should enable them (and their colleagues) to clearly distinguish in retrospect what they have thought a priori and what they have derived a posteriori from their study. In other words, preregistration makes it easier to separate confirmatory from exploratory analyses (see Nosek et al., 2018; Van den Akker, 2023).

There are different forms for a preregistration, which can be found on this link: https://osf.io/zab38/wiki/home/

If you have never written a preregistration, we recommend you to start with a simple form (e.g., the “Preregistration Template from AsPredicted”).

If you are already experienced with preregistration, we recommend you to use the validated OSF prereg or the detailed PRP Quant form.

Preregistrations can be carried out on the Open Science Framework (https://help.osf.io/article/158-create-a-preregistration) or the Leibniz Institute for Psychology (https://prereg-psych.org/index.php/rrp). There, you can upload your preregistration, anonymize it for peer review, and receive a link that you need to use in your research article.

Several journals even offer a preregistration with a peer review, which is called a “Registered Report”. If you are in the early stage of your project and you have a rough research idea, we recommend you to do a Registered Report! This format helps you with an external review of your research idea before you conduct the study and it is effective in counteracting publication bias. Moreover, PhD students are encouraged to use the “Peer Community In Registered Reports” platform, which offers standardized review procedures for Registered Reports.

Several journals even offer a preregistration with a peer review, which is called a “Registered Report”. This format helps you with an external review of your research idea before you conduct the study and it is effective in counteracting publication bias.

If you are in the early stage of your project and you have a rough research idea, we recommend you to do a Registered Report! Moreover, PhD students are encouraged to use the “Peer Community In Registered Reports” platform, which offers standardized review procedures for Registered Reports.

The availability of materials, data and analysis scripts facilitates the traceability (e.g., for reviewers) of the research process, the reproducibility of results, and facilitate a close replication of the study by other researchers. For even more transparency, it is also possible to document in open notebooks (e.g., project jupyter) which materials and methods were used to collect certain data.

Scientific repositories allow to deposit large amounts of data and commented code (e.g., via R or Python) online. As a starting point, we recommend checking out the Open Science Framework (OSF) or the curated ZPID to store (anonymized) data and materials easily online. For instance on the OSF, you have to register, then create a new project and upload your data via drag and drop. You can also set an embargo to make the data accessible only after some time. Soderberg (2018) offers a comprehensible guide on how to share data and materials step by step on the OSF.

A lot of journals put their research articles behind a “paywall” – hence, they cannot be accessed by other researchers or interested laypersons. However, there are multiple ways to make your article accessible (OA):

1) There are some OA publishers, who charge very low fees (also known as article processing charges) or no fees at all. Two relevant publishers for psychological research are PsychOpen and  ubiquity press.

2) The University Library may cover costs for publishing articles OA. Furthermore, the library also made publishing agreements with several journals, which you can check out here. If you find your preferred journal in the agreement lists, OA publishing costs are reduced or even fully covered.

3) You can use openly accessible preprint servers such as Zenodo or PsyArxiv to publish your articles. In most cases, the first submitted version of a research article (i.e., before the first peer review) can be stored on preprint servers without issues. And you can upload this early version even after publication. Details and exceptions for many journals can be found on Sherpa Romeo.

In research articles or chapters in your dissertation, you can make your OS practices visible by writing an “Ethics and Open Science Declaration”, where you indicate links to your preregistrations and accessible data and materials (see example here).

For your PhD thesis, we suggest to use the “Transparency and Open Science Statement” form, which you can find on the GOSI page. This form can be added to your dissertation and gives information on all OS practices on one glance.

The Doctoral School of Psychology aims to offer regular courses on research ethics and open science. These courses will be announced on the Curriculum page. Furthermore, the Deputy Head of the Doctoral School, Prof. Katja Corcoran is offering a graduate seminar “Open Science Practices” for Master and Doctoral students. If you are interested in OS practices more broadly, you are welcome to come to one of the GOSI meetings.

Contact

Head of Doctoral School
Universitaetsplatz 2/EG, 8010 Graz
Assoz. Prof. Mag.rer.nat. PhD Stephan VOGEL Phone:+43 (0)316 380 - 8478

Office hour:
on appointment

End of this page section.

Begin of page section: Additional information:

End of this page section.