Rehabilitation Research Design and Disability Center

xFACT Software Platform

Executive Summary

This website hosts information and resources related to the xFACT software platform, its original incarnation as OTFACT, and assessment applications that were developed to be usable in the xFACT platform.

xFACT is a software suite which operationalizes:1) the formulation, 2) the testing, and 3) the functioning of taxonomies based on TTSS (Trichotomous Tailored Sub-branching Scoring) system. For decades, Dr. Roger O. Smith’s research at the Rehabilitation Research Design and Disability (R2D2) Center has focused on assessments that measure the functional performance of individuals and the accessibility of environments, computers, learning materials, and products. The xFACT software platform helps an evaluator perform a comprehensive assessment efficiently and conveniently. It displays xFACT based taxonomies in a hierarchical manner and enables a user to score those taxonomies.

The xFACT software suite consists of three components, the Validator, Relevator, and Utility. The xFACT based taxonomies are all conceptualized, if not actually operationalized, to work in the xFACT platform based on the concepts of Trichotomous Tailored Sub-branching Scoring (TTSS) and the Taxonomy of Categories.

xFACT History

The historical foundation of xFACT emerged from a software package created for the field of occupational therapy developed during 1980s and 1990s named OTFACT. OTFACT was initially supported The American Occupational Therapy Foundation and Apple Computer, Inc.

View summary of OTFACT history

Development of the xFACT platform

Computerization of the pencil and paper version of OTFACT in the early 1990s led to development of the current xFACT software platform. Significant advancements in the xFACT software platform have recently been implemented to meet the needs of the R2D2 Center and others who are developing TTSS-based taxonomies that address an increasing need for more comprehensive and efficient assessments of functional performance of individuals, of assistive technologies, and of universal design that aims to improve accessibility and active participation in community for people with disabilities. Over more than two decades, the R2D2 has produced numerous different taxonomies. Efforts to validate and ascertain the usability of newly developed assessments led to advancements of the xFACT software platform, which now, in addition to functional assessments, also provides the xFACT Content Validator, the xFACT Relevator and an xFACT Taxonomy Editor to support the development of xFACT taxonomies.

Master Bibliography

This Master Bibliography includes all OTFACT and xFACT based taxonomies and all theses, dissertations, projects, publications, and presentations related to OTFACT and xFACT-based projects, as well as references to the origin, development, and use of the xFACT software platform.

Trichotomous Tailored Sub-branching Scoring (TTSS)

TTSS is a trichotomous scale used by xFACT based applications to score items in the taxonomy. TTSS is defined by three cardinal features: 1) trichotomous, 2) tailored, and 3) sub-branching, which make the TTSS scale a unique and efficient way to perform comprehensive assessments in a relative very shorter duration (Smith, 1999). Brief descriptions of the three features are as follows:

1. Trichotomous Scoring. All items are scored using a trichotomous scale; 0, 1, 2. A score of 0 is given if none of criteria are met for that particular question, and a score of 2 is given if a client/subject meets all of the criteria in the category. Perhaps of most importance are that score of 1 is given for all conditions where some of the criteria for a given item are met while some criteria remaining unaccomplished (Smith, 1999). More detail on the importance of these scores follows in section 3.
2. Tailored. In addition to the three responses, two additional responses, “not-applicable” and “not examined” are also provided for irrelevant questions to customize the assessment relevant to individuals in their particular contexts. The TTSS omits all the items selected as not applicable and not examined from the total scores. The results obtained from an assessment are comprised of the items relevant to the individual, thereby tailoring the assessment according to an individual’s particular needs and contexts (Smith, 1999).
3. Sub-branching. Sub-branching is a unique feature of a TTSS-based taxonomy, which allows for detailed analysis, yet also allows for a quick and efficient scoring process. If a score of 0 or 2 is given for an item, the TTSS process moves to the next branch of the same level within the taxonomy. If a score of 1 is given, the taxonomy opens sub-branches, which allow for a more detailed analysis (Smith, 1999).

Taxonomy of Categories (Category Framework)

The taxonomy of categories (Category Framework) consists of the large item set of a taxonomic assessment. The categories for an assessment should be created with a sound theoretical background of the construct for which the assessment was created, and based on past literature and other existing assessments. The categories should be as inclusive of the construct as possible. Categories should be organized in a hierarchical manner. It should be possible to breakdown each of the major categories into underlying concepts to best capitalize on xFACT based taxonomies.

The taxonomy of categories can be created either in an Excel worksheet and imported into xFACT, or created directly in the xFACT Utility. The taxonomy of questions is then organized into a branching TTSS organization.

Organization of a Taxonomy

Two analogies can be used to describe the branching organization of a taxonomy. The first depicts the taxonomy as a tree, which branches multiple times, until terminal “leaf” items appear. While the tree analogy describes the format, it does not handle the multi-role aspect of taxonomy items. A better analogy is that of a family. High level items are described as “Parents” who have “Children”. Parent items become “Grandparents” as their children have children items. Children items on the same level are considered “Siblings”. This analogy, however, does not describe well the terminal items, and so the two analogies of trees and families often become intermixed.

Taxonomies are organized in a hierarchical fashion. Depending on the response, the item can proceed directly to the next item of the same level, or branch into more detailed items which can be seen as members of the higher category. Branching continues until the terminal items are reached at the end of each branch. The taxonomy then proceeds to the next item, which is usually on a higher level.

While describing the branching system sounds confusing, it is actually cognitively easier than many traditional temporal computer aided testing (CAT) methods. In the TTSS system, the relationships of items remain visible, and allows for the user to go back to previous items at any time.

The xFACT based taxonomies takes advantage of a static list and dynamic sequencing to be able to display branching in a simple manner. The xFACT taxonomies uses spatial indentation (e.g. Roman, capital, small letter format in the sequential order to indicate different levels), signs (a plus sign indicates more categories or questions to follow), and other features such as ability to move back and forth to make it user-friendly and cognitively easier to comprehend.

Thus, xFACT based taxonomies, in conjunction with TTSS, allows for a large set of items to be answered efficiently, yet with as much detail as may be needed.

In order to ensure that the taxonomy is useable, understandable, and efficient, rules of writing taxonomy items and structuring the branching system have been established and are delineated below.

Rules for Creating Taxonomies: TTSS

Categories

• Parent categories should have a
• There should be 12 or fewer siblings under a parent category. If there are more than 12 siblings, create another parent category and split the siblings into the appropriate categories.
• Children/siblings are mutually exclusive.
• Verb tense and phrasing should be consistent for all categories (e.g., first person, active verb).
• Punctuation should be consistent (e.g., use periods or not at the end of sentences/phrases).
• Capitalization should be consistent.
• Lengths of categories should be short, and the acceptable length is based on the level of branching in the outline. (i.e., all category text should be completely visible when the xFACT platform is at its’ smallest allowable size).

Definitions

• Definitions should be short but complete.
• The parent definition should include the context and details of all the children categories below if the parent level can be scored (so that raters who score at the parent level know the details of what they are scoring).
• Definitions of the children do not need to repeat major concepts defined for parents.
• Definitions should be written at an appropriate reading level.

Responses

• Minimize the number of different response sets used for the taxonomy to make scoring cognitively simple.
• Use positive phrasing.
• Make sound decisions for using N/A versus Not Examined versus Continue.
• N/A means the item falls out of the data set and is not included in the final score.

Continue to XFact Software Development Suite >>