Proto-Genie
The word "Proto" in ProtoGenie™ is shorthand for "protocol" which refers to the plan of a scientific experiment or treatment." A protocol is a set of instructions to guide the conduct of an experiment. ProtoGenie automates protocols so that the computer will administer and record experimental sessions effectively, accurately, and reliably. "Genie" is the name of the mythical spirit that looks after its master. It is also French for "genius." Both are light-hearted metaphors for the way ProtoGenie helps the researcher overcome obstacles and makes good things happen.
ProtoGenie PGWorld
PGWorld will be a worldwide network or “research environment” of researchers brought together around ProtoGenie online to share software and information and to reduce dependence on software engineers and costly fixed applications. The Internet makes this rich research environment possible. It is designed to grow and evolve with universal access and open source tools. A central principle is the conservation of software effort through the recycling of protocols. In other words, ProtoGenie treats protocols as research CAPITAL to be exploited, not wasted. PG World makes the application of this principle possible.
Research Methods
Research methods are categories of terminologies, strategies, and techniques that are used to conduct research. This version of ProtoGenie supports five such categories or methods: Classical & Quasi Experiments, Clinical Trials, Laboratory Trials, Survey Research, and Observational Research. (See below)
Classical & Quasi Experiments
Classical experiments seek to measure the causal effects of treatment variables on response variables through random assignment of subjects to experimental groups and through the manipulation of the treatment variables; sometimes referred to as “laboratory experiments” because of high degree of control over settings; commonly used in psychology and related disciplines and in clinical settings, law and other professions. Studies that do not permit high levels of control are generally known as “Quasi-experiments” and are generally used in field settings such as schools and other institutions; classical experiments create situations as close to real as possible; quasi-experiments compensate for lack of controls through matching, placement and withdrawal of treatments, and statistical analysis.
Clinical Trials
clinical trial is a study that follows selected individuals forward in time from a pre-set baseline, some receiving an intervention and some not; typically measures the effects of medical interventions, including therapeutic agents, devices, regimens, and procedures; most commonly used in medical, pharmaceutical, and public health research; a major part of design is usually the provision of mechanisms and procedures for maximizing and assessing “compliance,” as in taking a medication daily in the prescribed amount at the prescribed times.
Laboratory Trials
Laboratory trials typically involve repeated presentations of stimuli (interventions or treatments) and measurements (trials); often using equivalent materials to control the effects of memorizing; often presented on visual and auditory devices; commonly used in vision, cognitive, and human performance research; also used in materials research and testing; examples include testing the effects of colored text and backgrounds on reading speed and comprehension and testing the effects of distractors on target detection and recognition; an industrial example might test the effects of variable temperatures on the properties of a material or product.
Survey Research
Survey research is the study of attitudes, beliefs, and behavior of people and their settings through questionnaires administered by mail, handouts, personal and, telephone interviews, and the Internet; ranging from one question polls to large-scale studies; sometimes employing panels and time samples; frequently used in national and local studies of political and economic attitudes and reported behavior; no conscious attempt to intervene to determine causality; used to identify important variables, to increase understanding, and sometimes to promote a change through education.
Observational Research
Observational research is the systematic, usually first-hand observation of some event in progress, such as a classroom, gang activity, legislative body, or discussion group, usually guided by a check list of anticipated events or phenomena and spaces for entering unanticipated events or phenomena as they occur. An example of an application of ProtoGenie in an observational study in vision science is a systematic recording of observations of accident avoidance behaviors and confrontations in a busy corridor at a convention for blind people. To build a protocol using this method, select a typical design from below, or click here to go to the startup protocols page for other startup options.
ProtoGenie Protocol
A protocol is an online “program” or “script” created by researchers using ProtoGenie that defines and guides events in experiments, clinical trials, surveys, and other kinds of studies. These protocols are created online and executed online. New protocols are made by customizing existing protocols selected from hundred’s of typical research designs, protocols built for ProtoGenie, and protocols created by other ProtoGenie users. Also, new protocols are made by modifying a protocol from a researchers personal library of protocols. If desired, protocols can be created from scratch with ProtoGenie’s step-by-step assistance, contextual help, and help by topic and key words.
Protocol Descriptions
Protocol descriptions contain information about each protocol, including the title, a description of the study, the topic and field, keywords that identify the study, author’s name, protocol number, date created, date last modified, research method category, and status of the protocol (public or private). This information is entered when a protocol is created and it is attached to the protocol records in the database to enable effective searches for protocols that fit a user’s study.
ProtoGenie Composer (Also referred to as Editor)
ProtoGenie Composer is the central component of ProtoGenie that enables researchers to create and modify protocols to guide their studies. Actions include the specification of groups, treatment events, measurement events, control events, and sequencing.
ProtoGenie Database
The ProtoGenie database contains ProtoGenie protocols that have been completed or are in progress and results from executing protocols in the form of data files. Protocols are retrieved through users’ personal library lists, by number, or by keyword searches.
Startup Protocols
There are four sources of protocols from which to choose a startup template: Blank protocol, Personal protocols, Typical Design Protocols (under development), and 3rd party protocols. See below.
Blank Protocol
The first alternative startup protocol is a Blank protocol. This means that the input fields for defining groups and specifying session events are presented but are completely empty. This choice over existing protocols is mostly a matter of preference and objectives.
Personal Protocols
The second startup alternative is a protocol from your own personal library of ProtoGenie protocols, assuming of course that you have previously used ProtoGenie to support your research and have saved them in the ProtoGenie database.
Typical Designs Startup Protocols (Under Development)
The third startup alternative is a protocol from a library of typical designs for the method selected. For example, a typical design for classical control group experiments is the popular pretest-posttest control group design in which one group receives a treatment with measurements before and after while the other group gets only the pre and posttest measurements. In these protocols, the locations in the specific designs where groups, measurements, and treatments are to be presented are indicated, but they are not specified. For example, where a treatment is to be defined in the protocol the label will simply be "Unspecified Treatment." The same goes for groups. See “Unspecified Groups and Events.”
General Library Of Protocols
The fourth startup alternative is a protocol from the ProtoGenie Archive consisting of pre-built and 3rd party protocols. These are accessed through a field and keyword search. The general idea is to find a protocol that someone has created that is close to your study.
Searching Protocols
You can search your own protocols or the general library of protocols by Title, ID, and Keyword. You can search by a specific method or all methods.
Measurement Events
In surveys, a measurement event is an action taken to determine attitudes, beliefs, knowledge, skills, memory, dispositions, and feelings. In experiments a measurement event is an action taken to determine whether something happened or changed in an experiment in response to a treatment (intervention or stimulus). Measurements generally involve things you want to learn about in your study and are usually referred to as "dependent variables" or "response variables" because they depend on or are responses to other factors that you want to explore, evaluate, or use to produce a change. Examples of dependent variables are "jury verdicts," "reading performance," and "reaction time." Variables must be "operationalized," meaning that we must specify exactly how they are measured. For example, a question might be asked or a blood pressure might be taken. These actions are measurement events.
Types of Measurement Events
Checkbox
Dropdown
Fill-in-the Blank
List Box
Matching
Multiple Choice
Ranking
Rating Scale
Short Answer
Sliderbar
True-False
Unspecified Measurement
Treatment Events (also Intervention and Stimulus Events)
A treatment event is an action taken to manipulate something in your experiment to study its effects on a dependent variable. For example, you might be interested in studying the effects of drinking coffee on reading performance. Treatment variables are called "treatments," "interventions," "change variables," and "stimuli," depending on the nature and setting of the study. The statistical term for these variables is generally "Independent variables," although this may be deceiving since variables are seldom independent of other influences. Treatment variables can be categorical in measurement as in "gender" or continuous, as in "letter contrast." In statistical analysis, categories are generally called "levels," and these are often treated as separate treatments. For example, one might investigate and compare the effects of blue, green, gray, and no filters on reading speed. The planned manipulation of an independent variable is a treatment event.
Types of Treatment Events
Administered by Experimenter
Administered by Machine
Administered by Other Person
Administered by Subject
Administered on Video
External Stimulus
Image Stimulus
Text Stimulus
Unspecified Treatment Event
Intervention Events (also treatment and stimulus events)
An Intervention Event is an action taken to manipulate or change something in an experiment to study its effects on an independent variable. The term “intervention” is most often used in clinical trials research, as opposed to “treatment” or “stimulus.”
Types of Intervention Events
See “Types of Treatment Events” above
Stimulus Event (Also Treatment and Intervention Events)
A Stimulus Event is the presentation of a visual or other sensory object to an action taken to manipulate or change something to study its effects on an independent variable. The term “stimulus” is most often used in clinical trials research, as opposed to “treatment” or “stimulus.”
Types of Stimulus Events
See “Types of Treatment Events” above
Support Events
Unlike treatment or measurement events in a protocol, support events are events strategically placed to guide the conduct of the study. Some support events are simply screens that instruct subjects or experimenters or alert them to some impending action. Also, support events are used to organize and schedule events in the study. Support event types are as follows:
Guidance Event
Branching Event
Calendarize Event
Event Iterations
Event Array
Study Context
Unspecified Event
Randomized Event
Goodbye Event
Experimental Groups
Groups of subjects are often defined for controls and comparisons in experiments. Subjects may be assigned to different groups based on subject attributes or they may be randomly assigned to groups that will receive different treatments. A special category of treatments is “no-treatment.” Groups given no-treatments are generally called “control groups.” Consequently, the major categories of experimental groups are Treatment Groups and Control Groups.
Treatment Groups
A common use of experimental groups is to assign treatments to separate groups. If assignment is random, then the groups will be statistically independent of the treatment variables, thereby making it easier to interpret the specific effects of the different treatments. Another use of groups is to compare the effects of a treatment or treatments in one population to the effects in another. For example, one might be interested in the question of whether there are differences between the memories of people diagnosed as dyslexic and "normal" people and whether background colors affect memory differently across groups.
Control Groups
A common design for experiments calls for random assignment of subjects to a group that does not receive a treatment. Groups are also formed in multiple treatment designs when the design calls for a group for every intersection of treatments (including no-treatments). These are generally called “factorial” or block designs.
Null Event (Also called a place holder Event)
Unlike a treatment event or a measurement event where something is supposed to happen when the protocol is executed, nothing happens when a Null Event is encountered. The purpose of inserting null events is to fill the time intervals in which other groups are presenting measurements or treatments - thereby aligning measurements and treatments across groups and assuring that measurements and treatments are presented at the appropriate time in each group. In this sense, they are “place keepers.” If it is not important that treatment or measurement events in one group are “aligned” (taken at the same time as treatment or measurements in other groups), then null events are not needed.
Unspecified Groups and Events
Unspecified groups and events are generally used in the creation of “Typical Design Protocols” (under development) for use as startup protocols where the aim is to indicate where groups and events occur by design. For example, a typical experimental design is the Solomon Four-Group Design. Unspecified labels are used to indicate a single treatment, two measurements, and four groups. Also, unspecified treatment and measurements are placed on a timeline for each group as specified by the Solomon Four-Group design. The user’s task is to specify each unspecified group and event. Besides use in typical design startup protocols, it is conceivable that some researchers would insert unspecified groups or events in a protocol to remind themselves that they want to add a group or event that they have not yet defined.
Organizing and Scheduling Research Events
Creating protocols involves the specification of WHAT events are supposed to take place in a study, but creating protocols also involves the specification of exactly WHEN events should take place. This includes the specification of the order events should take place, the specification of time intervals between events (or on the calendar), and the specification of contingencies that change the order or timing of events in response to input. For details, click here.
Dr. Lawrence H. Boyd, President, Pasadero, Inc., Tucson, Arizona
Home Support System Requirements Contact Communication Gap Challenge Projects PG & Education Applications Research Methods Origins/History Key Concepts/Definitions Terms Of Use Privacy Policy Site Map
© 2000 - 2006 Pasadero, Inc. All Rights Reserved.
ProtoGenie™
United States and Foreign Patents Pending
 |
 |
 |
|
||
|
|
||
 |
|
 |
|
||
© 2000 - 2007 Pasadero, Inc. All Rights Reserved.
ProtoGenie™
United States and Foreign Patents Pending