Measurement 
Jacqueline Corcoran 
Virginia Commonwealth 
University 
From: Social Work Research Skills Workbook, Corcoran & 
Secret, 2013, Oxford 
http://www.jacquelinecorcoran.com/
Measurement 
 Concepts: ideas 
 Operationalized: how concepts are 
translated into numerical terms
Defining Variables 
 Independent: 
 What we want to study for its impact 
 In intervention research: what we 
manipulate 
 Dependent: 
 The result, outcome 
 Changes as a result of the action of the 
independent variable
Levels of measurement 
 Nominal 
 categories, exhaustiveness and mutually exclusivity 
 Questions: 
 Does this fall into a category? 
 What category does this fall into? 
 Ordinal 
 can be logically rank-ordered 
 differences between ranks can’t be quantified 
 Questions 
 Can I place these items in an approximate order? 
 Interval 
 the distance separating attributes has meaning but no real zero 
 Questions: 
 Can I place this on a scale that is very exact and precise? 
 Ratio 
 the distance separating attributes has meaning and a real zero 
 Can I place this on a scale that is very exact and precise, AND has an 
absolute value?
Assigning Values to Variables 
 Nominal: numerical values are assigned 
arbitrarily 
 Ordinal: use numbers that symbolize the 
levels 
 Interval and ratio: use the actual value.
Treatment Fidelity 
 refers to how faithfully we have 
implemented the intervention 
 Internal validity 
 Includes proper training and 
supervision and examination of their 
work in sessions to ascertain that it 
sufficiently meets the standards and 
tenants of that particular treatment as 
it was designed. 
 Manualized treatment (treatment that 
follows a strict protocol)
Reasons to use standardized 
measures 
 You will not be able to standardize a new 
instrument 
 Scoring and their interpretation 
 Too many statistical tests 
 Not re-inventing the wheel
Ways to find free measures 
 Find out if your agency has purchased a 
measure for its use. 
 Contact the creator of a measure and 
request permission to use the measure for 
educational purposes. 
 Bibliographies 
 Fischer and Corcoran (2007a, 2007b) 
 Corcoran and Hozack (2010) 
 For strengths-based measures and how to 
access them, see Early and Newsome 
(2004).
Agency Records 
 Example 8.4 for key outcomes 
 Certain types of agencies routinely collect 
certain information that can be used for 
student projects

Measurement

  • 1.
    Measurement Jacqueline Corcoran Virginia Commonwealth University From: Social Work Research Skills Workbook, Corcoran & Secret, 2013, Oxford http://www.jacquelinecorcoran.com/
  • 2.
    Measurement  Concepts:ideas  Operationalized: how concepts are translated into numerical terms
  • 3.
    Defining Variables Independent:  What we want to study for its impact  In intervention research: what we manipulate  Dependent:  The result, outcome  Changes as a result of the action of the independent variable
  • 4.
    Levels of measurement  Nominal  categories, exhaustiveness and mutually exclusivity  Questions:  Does this fall into a category?  What category does this fall into?  Ordinal  can be logically rank-ordered  differences between ranks can’t be quantified  Questions  Can I place these items in an approximate order?  Interval  the distance separating attributes has meaning but no real zero  Questions:  Can I place this on a scale that is very exact and precise?  Ratio  the distance separating attributes has meaning and a real zero  Can I place this on a scale that is very exact and precise, AND has an absolute value?
  • 5.
    Assigning Values toVariables  Nominal: numerical values are assigned arbitrarily  Ordinal: use numbers that symbolize the levels  Interval and ratio: use the actual value.
  • 6.
    Treatment Fidelity refers to how faithfully we have implemented the intervention  Internal validity  Includes proper training and supervision and examination of their work in sessions to ascertain that it sufficiently meets the standards and tenants of that particular treatment as it was designed.  Manualized treatment (treatment that follows a strict protocol)
  • 7.
    Reasons to usestandardized measures  You will not be able to standardize a new instrument  Scoring and their interpretation  Too many statistical tests  Not re-inventing the wheel
  • 8.
    Ways to findfree measures  Find out if your agency has purchased a measure for its use.  Contact the creator of a measure and request permission to use the measure for educational purposes.  Bibliographies  Fischer and Corcoran (2007a, 2007b)  Corcoran and Hozack (2010)  For strengths-based measures and how to access them, see Early and Newsome (2004).
  • 9.
    Agency Records Example 8.4 for key outcomes  Certain types of agencies routinely collect certain information that can be used for student projects