154 research outputs found
Svenska skolreformer till mer skada än nytta [Elektronisk resurs]
No full textNy internationell forskning visar att många svenska skolreformer inte föregåtts av utvärderingar utan baserats på mer eller mindre rimliga teorier som lanserats som sanningar. Åldersblandade klasser, problemorienterad undervisning, läxor och skriftliga test är metoder som är skadliga eller bara ger måttlig effekt på inlärningen, skriver Knut Sundell, Sven Bremberg och Martin Karlberg.</p
Svenska skolreformer till mer skada än nytta
No full textNy internationell forskning visar att många svenska skolreformer inte föregåtts av utvärderingar utan baserats på mer eller mindre rimliga teorier som lanserats som sanningar. Åldersblandade klasser, problemorienterad undervisning, läxor och skriftliga test är metoder som är skadliga eller bara ger måttlig effekt på inlärningen, skriver Knut Sundell, Sven Bremberg och Martin Karlberg
Supplemental Material, RSWP633297_Supplementary_Material - Trends in Methodological Quality in Controlled Trials of Psychological and Social Interventions
No full textSupplemental Material, RSWP633297_Supplementary_Material for Trends in Methodological Quality in Controlled Trials of Psychological and Social Interventions by Haluk Soydan, Knut Sundell, and Elizabeth Åhsberg in Research on Social Work Practice</p
Utvärdering av importerade interventioner
No full textBokinformation: Varje år berörs hundratusentals personer av psykosociala och pedagogiska interventioner. Det handlar exempelvis om behandling av missbruk, insatser för utsatta barn och läsundervisning i grundskolan. Endast undantagsvis har dessa interventioner utvärderats, vilket innebär att vi inte vet vilken effekt de har eller om de rentav kan skada. I antologin Att göra effektutvärderingar beskrivs hur man genomför effektutvärderingar, som ger kunskap om hur interventioner fungerar. Boken ger detaljerad och praktisk vägledning som är till nytta för alla som är intresserade av att lära sig mer om effektutvärderingar – både forskarstuderande och forskare som redan ansvarat för effektutvärderingar. Ämnen som tas upp är bland annat forskningsetik, studiedesign, mätinstrument, urval och rekrytering av undersökningspersoner, datainsamling, ekonomiska analyser, dataanalys och tolkning samt rapportskrivning. Bokens författare har lång erfarenhet inom området och representerar olika discipliner såsom psykologi, folkhälsovetenskap, socialt arbete, pedagogik, statistik, epidemiologi, sociologi, evolutionsbiologi, medicinsk vetenskap och datavetenskap. Redaktör för boken är Knut Sundell, docent i psykologi och socialråd på Socialstyrelsen, som har lång erfarenhet av att utvärdera interventioner inom förskola, skola och socialtjänst.</p
Supplemental Material, PRISMA_2009_checklist - Treatment Foster Care Oregon for Delinquent Adolescents: A Systematic Review and Meta-Analysis
No full textSupplemental Material, PRISMA_2009_checklist for Treatment Foster Care Oregon for Delinquent Adolescents: A Systematic Review and Meta-Analysis by Therese Åström, Martin Bergström, Kickan Håkansson, Ann Kristine Jonsson, Christian Munthe, Ingegerd Wirtberg, Johanna Wiss and Knut Sundell in Research on Social Work Practice</p
Supplemental Material, Supplement_bilaga-4-sokstrategier - Interventions in Foster Family Care: A Systematic Review
No full textSupplemental Material, Supplement_bilaga-4-sokstrategier for Interventions in Foster Family Care: A Systematic Review by Martin Bergström, Marianne Cederblad, Kickan Håkansson, Ann Kristine Jonsson, Christian Munthe, Bo Vinnerljung, Ingegerd Wirtberg, Pernilla Östlund, and Knut Sundell in Research on Social Work Practice</p
Scientific productivity in social science disciplines related to health and caring
No full textThe problem to be addressed
Timely and impactful dissemination of knowledge is essential, and this is arguably even more important when it informs practises that have a direct impact of people’s health and well-being (Sundell, 2010). It has been documented that disciplines differ substantially in the number of journal articles that each researcher produces, although they all serve the health and social services, without rational cause (e.g., Olsson & Sundell, 2015). Almost all such studies are based on what is actually published, and are therefore not informed about the base frequency through knowing how many active researchers there are. Obviously, a discipline with 100 researchers should produce one-third of a discipline with 300 researchers, if all researchers are equally productive. Moreover, authors who do not publish anything at all in the databases used are not noticed, although they should contribute to a lower mean productivity for their discipline. To provide a fair comparison of disciplines, we therefore identified the whole population of researchers in Education, Sociology, Social Work, Psychology, Public Health, and Nursing and Caring Science, and measured the mean numbers of publications and citations across all researchers in each discipline for the period 2000-2009 (Madison & Sundell, 2022).
With these representative data of the population of researchers in Sweden we were also able to address the well-documented sex difference in scientific productivity. The data showed that, for all six disciplines, the vast majority of women and men are equally productive, and constitute similar proportions across low- and medium levels of productivity (10th-80th percentile). Above the 80th percentile, however, the proportion of males increases with the level of productivity (Madison & Sundell, 2023).
The present pre-registered project aims to replicate these two previous studies, and to compare the outcomes across the 12 years that have elapsed. To this end, we will again identify the whole population of researchers in Sociology, Social Work, Psychology, Public Health, and Nursing and Caring Science, but not Education, and obtain their numbers of publications and citations for the period 2012-2021. Researchers are defined as faculty at academic institutions that are at least associate professors (docent). In contrast to our previous study, which analysed productivity for the whole population, we will now draw randomized stratified samples optimised for the various research questions. This will increase statistical power and decrease the workload. Specifically, we will create one sample optimised for comparing disciplines, having the same proportion of docent and professor, and male and female researchers, as in the whole population. Another sample will be optimised for comparing male and female researchers at both docent and professor academic levels within each discipline, having equal sample sizes of 150 individuals for each of these four groups (30 for each discipline) but otherwise the same proportion of researchers from institutions with various levels of eminence and productivity as in the whole population.
Methods
Researchers are identified through the institutions at which they are employed. We will contact heads of departments at all institutions in Sweden that contain the target disciplines through email, and ask them to provide a list of research staff, as defined above. This includes all departments of Sociology, Social Work, and Psychology, and various departments relevant for Public Health and Nursing and Caring Science, which may belong to either Medicine or Social Sciences.
Then, randomized, equally-sized samples will be created to optimize analyses comparing disciplines and comparing males and females, provided that the population N varies across disciplines or sex.
The full record of publications for each researcher will then be downloaded from the Web of Science, from which the numbers of publications and citations for each year in the period 2012-2021 is extracted.
Analyses will involve numbers of publications, citations, and numbers of publications per year as dependent variables. Independent variables are discipline, sex, academic rank (associate or full professor), and academic age, defined as the number of years since the start of a consistent publication record or a significant bout of publications.
At the time of writing this pre-registration, we have gathered the researcher names from the institutions, but we have not compiled them in a data matrix, extracted their publication records from the Web of Science, nor performed any analyses.
Hypotheses
(H1) Productivity (i.e. numbers of publications, citations, and numbers of publications per year) is greater for 2012-2021 than for 2000-2009, across sex and discipline. This is because some disciplines publishes more in peer-reviewed journals.
(H2) As found for 2000-2009 (Madison & Sundell, 2022), productivity remains highest for Public Health and Nursing and Caring Science, and next highest for Psychology, followed by Sociology and Social Work.
(H3) Productivity differences between disciplines is smaller for 2012-2021 than for 2000-2009. This is because we predict that Sociology and Social Work has begun publishing more in peer-reviewed journals, while Psychology, Public Health and Nursing and Caring Science remain about the same.
(H4) The productivity difference between women and men is smaller in 2012-2021 than in 2000-2009, when males published 20-162% more journal articles, received 27-308% more citations, and received -0.5-224% more citations per publication, depending on discipline (Madison & Sundell, 2023). This is predicated on the fact that the proportion of females in faculty has increased in the last decades, to the effect that their academic age should be essentially the same as men’s in 2012-2021 even if it was younger in 2000-2009. For example did the proportion of female full professors in Sweden increase by about 10 percentage units from ~20% to ~30% from 2008 to 2014 (Madison & Fahlman, 2020).
(H5) A substantial productivity difference between women and men prevails in 2012-2021 above the 80th percentile, where the proportion of males increases with the level of productivity. This is predicated on the greater male variability hypothesis, which states that psychological traits that predict academic precociousness and productivity varies more in males. Therefore, the males in the strongly selected population that survives the gate-keepers of the academic system will contain a much larger proportion of exceptionally talented males than exceptionally talented females. Hence,
(H6) variability in productivity (both number of publications and citations) is larger in males than in females, as indicated by a variance ratio (the variance amongst males divided by the variance amongst females) greater than 1.0 (Madison & Sundell, 2023).
(H7) The sex difference in productivity above the 80th percentile is larger in 2012-2021 than in 2000-2009. This is predicated on the fact that if the proportion of the group with smaller variability increases (females) then the proportion of exceptionally talented individuals within this group will decrease as individuals have to be selected from a more rapidly dwindling pool of such individuals. In contrast, the smaller proportion of males will be selected from a larger and increasing pool of exceptionally talented males.
(H8) H1 through H7 holds even when controlled for academic rank. H5 and H6 holds even when controlled for academic age.
These hypotheses will be evaluated through multivariate statistical analyses, primarily comparing group means while maybe controlling for possible confounding variables. For H6 the variance ratio will be used, computed as the ratio of variances based on individually log transformed data.
References
Madison, G. & Fahlman, P. (2020). Sex differences in the number of scientific publications and citations when attaining the rank of professor in Sweden. Studies in Higher Education, 46, 2506-2527.
Madison, G. & Sundell, K. (2022). Numbers of publications and citations for researchers in fields pertinent to the social services: a comparison of peer-reviewed journal publications across six disciplines. Scientometrics, 127, 6029-6046.
Madison, G. & Sundell, K. (2023). Sex differences in scientific productivity and impact are largely explained by the proportion of highly productive individuals: a whole-population study of researchers across six disciplines in Sweden . Studies in Higher Education, 49, 119-140.
Olsson, T. M. & Sundell, K. (2015). Research that guides practice: Outcome research in Swedish PhD theses across seven disciplines 1997-2012. Prevention Science, 17, 525-532.
Sundell, K. (2010). Internationella publikationer och citeringar under perioden 2000–2009 hos svenska professorer och docenter inom folkhälsovetenskap, omvårdnadsvetenskap, pedagogik, psykologi, socialt arbete och sociologi Socialstyrelsen
En bok om barnavårdsutredningar
No full textKnut Sundell & line Egelunds bok "Barnavårdsutredningar. En kunskapsöversikt'', recenseras här av docent Dagmar Lagerberg, verksam vid Barnhälsovården, Akademiska sjukhuset i Uppsala. Boken är utgiven av Socialstyrelsen: Centrum för utvärdering av socialt arbete och Förlagshuset Gothia 2000. 208 sid
Forskning om funktionshinder pågår 2009:1
No full texts. 2. Hur ska hjälpmedlet användas? – alltför stort teknikfokus i tillgänglighetsforskning s. 3. Läs tidningen på webben! s. 4. Hörselrehabilitering under lupp s. 6. Forskarporträtt: Snårigt liv efter olyckan s. 8. ADHD-forskning: Den svårfångade intelligensen s. 9. Hallå där! Knut Sundell s. 10. Läsvärt s. 10. Avhandlingar s. 12. Kalendern Felaktig ISSN i tidningen: 1103-1956. Den rätta är 2000-1789</p
Forskning om funktionshinder pågår 2009:1
No full texts. 2. Hur ska hjälpmedlet användas? – alltför stort teknikfokus i tillgänglighetsforskning s. 3. Läs tidningen på webben! s. 4. Hörselrehabilitering under lupp s. 6. Forskarporträtt: Snårigt liv efter olyckan s. 8. ADHD-forskning: Den svårfångade intelligensen s. 9. Hallå där! Knut Sundell s. 10. Läsvärt s. 10. Avhandlingar s. 12. Kalendern Felaktig ISSN i tidningen: 1103-1956. Den rätta är 2000-1789</p
- …
