What Works in Parenting Support: Online Database
Policy Research Bureau

* What Works in Parenting Support? Main report.
* What Works in Parenting Support? Summary report.

* Search the database
* Glossary

* Return to PRB Home Page


Descriptive information

These fields give descriptive information about the intervention that is the subject of the study.

Refers to the reference number of the study. Shows main outcome category (1a, 1b etc) and specific study number (1,2,3,4 etc) within that outcome category. These ID numbers are also used to identify and order the outcome commentaries in the Review publication. An explanation of the outcome categories is provided at http://www.prb.org.uk/wwiparenting/outcomes.html.
Main outcome category Searchable
Refers to the main outcome of the intervention as reported by the study (see the explanation of the outcome categories). Note that outcomes concern the actual outcome reported by the evaluators of the intervention rather than the intended outcome, as these can be quite different in practice.
Author(s) Searchable
Author(s) of the study.
Study Searchable
Name of the journal article, book chapter or other publication in which the evaluation was reported. This includes the year of publication. The 'author(s)' and 'study' fields combined give the full bibliographic reference for the study.
Exemplar status?
Where there are multiple trials or evaluations of a given intervention, up to three 'exemplar' evaluations have been entered into the grid rather than numerous evaluations of the same intervention.
Beneficiary or target population/group
Whether intervention is universal or targeted, whether aimed at parents or parents and children and including target group where relevant e.g. fathers, foster, grandparents, step parents, ethnic or cultural minority, low income, teenage, other.
Preventative/Therapeutic Searchable
Whether preventative or therapeutic.
Individual/Group work Searchable
Whether individual (one to one) work or group work. 'Group' includes school based programmes.
Format of the intervention
'Format' refers to different components or strands of the delivery, e.g. formal 'parent education' classes & courses, incl. parent training and skills building; advice and information-giving interventions incl telephone helplines & web-based services; home visitation by professionals, volunteers, family aides, peer supporters; and therapy or counselling for families and individuals.
Age of child
Numbers of parents, children or families using the service, and period used for (week, month, year, etc).
Intensity and duration
Number, length and frequency of sessions (e.g '2 x 2 hrs sessions each week, for three months').
Referral routes
Whether access to service is by self-referral, or by agency; whether attendance is voluntary or mandatory, etc, where reported.
Includes information on funding sector (e.g. statutory/voluntary); providing agency (e.g. health, social services, education, youth or criminal justice, leisure services, voluntary organisation); and staffing (e.g professionals /paraprofessional/peers etc), where reported.
Overall and specific objectives
Overall aims of the intervention, and specific outcomes for parents or children that are targeted within the broader aims, where reported.
Risk/protective factors
The risk and protective factors that the intervention aimed to decrease or bolster, where reported.
Information on costs and benefits of the intervention, including cost effectiveness or cost-to-benefits ratio, where reported.
Country Searchable
Country in which the intervention and study are located.

Evaluative information

These fields include the scientific and evidential factors in respect of methodological characteristics and quality of the evaluation. They also detail the evidence regarding the effectiveness of the intervention.

Quantitative study methods
Includes information regarding the following characteristics, where reported:
  • design: RCT/ quasi-experiment/uncontrolled study
  • measurement points: the number of separate waves of data collection (including follow-up studies)
  • sample size: 'small' is <50; 'medium' is 50-100; 'large' is 100+)
  • mode of administration: by whom data were collected, e.g whether by self-report, investigator-based assessment, observation etc
  • measures: whether measures used were validated, unvalidated or a combination
Qualitative study methods
Includes information regarding the following characteristics, where reported:
  • sample size
  • mode of data collected, e.g whether by depth interviews, group discussions, observation etc
Score on SMS Searchable
Score on adapted version of the 'Scientific Maryland Scale' (Farrington, Gottfreson, Sherman and Welsh, 2002; see review for more details). Scored 1 to 5 for quantitative studies only, with higher score indicating better methodological robustness as follows: Level 1: Association between a prevention program and an outcome measure at one point in time Level 2: Includes pre- and post- intervention measures (i.e. measures at two points in time) , but with no control group Level 3: Pre- and post-intervention measures (i.e. measures at two points in time) and also treatment and control group Level 4: Pre- and post-intervention measures (i.e. measures at two points in time) and treatment and control group, and also control for other factors that influence outcome Level 5: RCT i.e. Pre- and post-intervention measures (i.e. measures at two points in time) and treatment and control group, with participants randomised to treatment and control group) Note that some studies could have more than one score because they use a range of methods, or use different methods to evaluate different components of the programme. In these cases we have indicated only the highest score - ie the score for the most rigorous component of the evaluation. Some studies use methods which the SMS cannot score - they have been given a score of 0.
Score on GAEQ
Score on 'Global Assessment and Evaluation of Quality' (Moran, Ghate, Van der Merwe, 2004, see review for more details). Scored from 0 to 6 for quantitative studies and 0 to 5 for qualitative studies, on basis of number of affirmative answers with higher scores indicating better methodological robustness, as follows:
Dimension Quantitative studies Qualitative studies
Measures/ data collection tools Standardised (ie validated) or well-designed measures Specified and standardised data collection tools (e.g. written topic guides, aide-memoirs etc.)
Sample representativeness Response rate reported Adequate representativeness of sample relative to analytic dimensions (in sense of cross-section, not statistical representativeness) e.g. not all 'volunteers'; not all one type of person when intervention is delivered to a range)
Sample size Adequate sample size in relation to conclusions drawn Adequate sample size in relation to conclusions drawn (especially re: sub groups: not less than n=5)
Analytic methods Appropriate statistical methods Proper data capture methods (ie tapes, notes etc) & appropriate and specified methods of analysis (e.g. grounded theory; content analysis; framework analysis; thematic etc)
Programme integrity Programme integrity or fidelity taken into account in conclusions (No qualitative equivalent)
Type of evaluation External or independent evaluation External or independent evaluation
Negative and Positive indicators
Where reported, this field gives information relevant to the implementation and delivery of the intervention, including indicators of lack of effectiveness as well as of successful implementation, under the following categories:
  • attendance & engagement
  • cultural sensitivity/appropriateness
  • successfully reached range of ethnic groups (where intended to be 'inclusive' programme)
  • user satisfaction & perception of 'helpfulness'
  • practical aspects (location & accessibility; cost to users; timing)
  • continuity & appropriateness of staffing
  • (inter)agency working issues (communication, shared values etc) cost effectiveness
Works, Promising, and Doesn't Work
This field gives information on the reported outcomes (positive and negative) of the intervention as reported in the study. They are listed under one of three headings depending on the findings of the study: Works, Promising, and Doesn't work in relation to any outcome reported in the study (not just the main outcome).

The headings are adapted from Farrington et al (2000), and are defined as follows:

Works: an intervention with at least one SMS level 3 to 5 evaluation showing statistically significant and desirable results, and the preponderance of all available evidence showing effectiveness

Promising: where level of certainty is too low to support generalisable conclusions, but where there is some empirical basis for predicting that further research could support such conclusions

Doesn't work: an intervention with at least one SMS level 3 to 5 evaluation with statistical significance tests showing ineffectiveness, and the preponderance of evidence supports the same conclusion.

Where studies report on multiple outcomes, the various outcomes are listed. Numbers in brackets after outcomes refer to the reference number for that outcome category, by which commentaries in the main report are ordered and where further discussion of the particular outcome can be found (e.g. 1b refers to educational outcomes; see the explanation of the outcome categories).


For enquiries about The POLICY RESEARCH BUREAU's past work,
email dghate@prb.org.uk.
(c) 2004 Policy Research Bureau and Arches Consulting Ltd
$Id: fields.html,v 1.3 2007/03/30 12:05:06 tom Exp $