Part 1: Non-Experimental versus Experimental Research Non-Experimental Research:
ID: 388781 • Letter: P
Question
Part 1: Non-Experimental versus Experimental Research
Non-Experimental Research: When a researcher’s goal is to understand how users interact with a product in their everyday environment or to determine general characteristics and/or preferences of the user group, non-experimental methods such as interviews, surveys, or observational research methods are often used. Interview and survey methods can be used to elicit information from users, such as their understanding of how the system works or their preferences about different design characteristics. Usually, researchers will use interviews when they want to obtain in-depth information about a small group of users and surveys when they want to obtain general characteristics (e.g., age, income, experience) and preferences from a large group of users. Observational methods are used when the researcher wants to understand the context in which human-machine interaction takes place in order to design products that fit the users’ needs.
Listed below are goals for different research projects. Indicate which of the three non-experimental methods described above would be most appropriate for each project and describe why.
Goal: To determine whether the users of a product consist mainly of high, middle, or low income families.
Goal: To determine how expert computer technicians would go about determining the causes of a system error.
Goal: To determine whether workers in the office actually adjust ergonomic chairs so that they are seated comfortably at different workstations.
Experimental Research: Many human-machine interactions take place in an environment where different variables could affect performance, making it difficult to determine the causal relations between variables. In an experiment, the basic idea is to control variables that are not of current interest and to isolate the critical ones. The researcher manipulates one or more variables, called independent variables (e.g., the size of a display), and measures their effects on behavioural or system variables, called dependent variables (e.g., the total time spent on a task). A good experimental design places emphasis on control of the environment to ensure that only the independent variables could systematically affect the dependent variables. With their emphasis on control, experimental methods provide an objective method for establishing facts and evaluating alternative possible explanations. Experimental methods are excellent at helping human factors professionals evaluate differences in performance of alternative system designs and determine the factors that contribute to better performance.
For the following experiments, (a) identify the independent and dependent variables, and (b) evaluate whether the researcher was successful at eliminating the effects of unwanted or confounding variables in order to make his/her conclusion valid.
The director of an assembly factory wanted to lower the cost of electricity by switching to a light bulb that produces dimmer lighting. Since the director is aware that the factory workers’ productivity may be affected by the dimmer lighting, he does not want to change the light bulbs if it will result in a decrease of productivity. To help the director make a decision, the researcher measured the number of parts that the factory workers were able to assemble over a two-day period. During the first day, the researcher randomly assigned workers to work in Wing A of the factory, which used the current light bulbs, or Wing B, which used the new light bulbs. On day two, the researcher had each group work in the alternative location (e.g., those originally assigned to Wing A now worked in Wing B). Workers assembled 4,000 parts in Wing A, but only 3,400 parts Wing B. Based on these findings, the researcher concluded that the dimmer lighting provided by the new light bulbs caused a decrease in productivity and, consequently, recommended that the director should not change to the new light bulbs.
Independent Variable(s):
Dependent Variable(s):
Is the researcher’s conclusion valid? Explain.
Last summer, the E-commerce merchant, HumanFactors.com, wanted to evaluate which of two versions of their Website would produce the most sales. The researcher decided to conduct an experiment in which the number of sales was recorded for version A of the Website during the 3-month period of July-September, and the number of sales for version B during the 3-month period of October-December. Results showed that there were 500 sales for version A and 2,000 sales for version B of the Website. Based on these results, the researcher concluded that version B of the Website was better at promoting sales than version A.
Independent Variable(s):
Dependent Variable(s):
Is the researcher’s conclusion valid? Explain
Part 2: The Heuristic Evaluation—A Usability Inspection Method
A heuristic evaluation is a usability method designed to help human factors professionals inspect a product with the intent of finding usability problems in the interface design that can be fixed and re-evaluated as the design process progresses. Unlike the user-based methods described earlier in this project, the heuristic evaluation is usually conducted by usability professionals to examine whether the product or interface complies with recognized usability principles called heuristics. Below are the most widely cited heuristics, called the “Top Ten Usability Heuristics,” developed by Jakob Nielsen and Rolf Molich in the 1990s.
1.Visibility of system status: The system should always keep users informed about what is going on, through appropriate feedback within reasonable time.
2.Match between system and the real world: The system should speak the users' language, with words, phrases and concepts familiar to the user, rather than system-oriented terms. Follow real-world conventions, making information appear in a natural and logical order.
3.User control and freedom: Users often choose system functions by mistake and will need a clearly marked "emergency exit" to leave the unwanted state without having to go through an extended dialogue. Support undo and redo.
4.Consistency and standards: Users should not have to wonder whether different words, situations, or actions mean the same thing. Follow platform conventions.
5.Error prevention: Even better than good error messages is a careful design which prevents a problem from occurring in the first place. Either eliminate error-prone conditions or check for them and present users with a confirmation option before they commit to the action.
6.Recognition rather than recall: Minimize the user's memory load by making objects, actions, and options visible. The user should not have to remember information from one part of the dialogue to another. Instructions for use of the system should be visible or easily retrievable whenever appropriate.
7.Flexibility and efficiency of use: Accelerators -- unseen by the novice user -- may often speed up the interaction for the expert user such that the system can cater to both inexperienced and experienced users. Allow users to tailor frequent actions.
8.Aesthetic and minimalist design: Dialogues should not contain information which is irrelevant or rarely needed. Every extra unit of information in a dialogue competes with the relevant units of information and diminishes their relative visibility.
9.Help users recognize, diagnose, and recover from errors: Error messages should be expressed in plain language (no codes), precisely indicate the problem, and constructively suggest a solution.
10.Help and documentation: Even though it is better if the system can be used without documentation, it may be necessary to provide help and documentation. Any such information should be easy to search, focused on the user's task, list concrete steps to be carried out, and not be too large.
Visit http://www.usability.gov/ and explore the website. Evaluate whether Usability.gov complies with Heuristics 1, 5, 8, and 10. Be sure to include a brief statement regarding the nature of the compliance or how the website violates the heuristic. (For example, if the site does not have a help section or any documentation, state: The website violates heuristic 10 because it does not provide a help feature or documentation about how to use specific features of the site). Include at least one screen shot of a website page that you used to evaluate compliance with one of the heuristics.
Visibility of system status:
Error prevention:
Aesthetic and minimalist design:
Help and documentation:
Explanation / Answer
Even as there are lots of forms of quantitative research designs, they mostly fall beneath certainly one of two umbrellas: experimental research and non-experimental research. Experimental research designs are what many lay-individuals feel of once they feel of study; they on the whole contain the manipulation of variables and random undertaking of participants to stipulations. A ordinary experiment could involve the assessment of a manage crew to an experimental team who receives a treatment (i.E., a variable is manipulated). When completed competently, experimental designs can provide proof for rationale and result. When you consider that of their capacity to verify causation, experimental designs are the gold-common for research in medicine, biology, and so on. Nonetheless, such designs can also be used within the tender sciences, like social science. Experimental study has strict specifications for manage inside the study design and for starting validity. These designs may also be very useful resource and labor intensive. Additionally, it may be rough to justify the generalizability of the results in an extraordinarily tightly controlled or synthetic experimental setting. However, if accomplished well, experimental study ways can lead to some very convincing and interesting results.
Non-experimental study, then again, can be just as intriguing, however you are not able to draw the identical conclusions from it as that you may with experimental research. Non-experimental research is probably descriptive or correlational, which means that you're both describing a main issue or phenomenon readily because it stands, or you might be describing a relationship between two or extra variables, all without any interference from the researcher. Because of this you do not manipulate any variables (e.G., change the stipulations that an experimental staff undergoes) or randomly assign individuals to a control or medication group. With out this degree of manipulate, you can't investigate any causal effects. Even as validity remains to be a drawback in non-experimental study, the worries are more in regards to the validity of the measurements, as a substitute than the validity of the results.
So, how do you decide on between the two designs? This is determined by your matter, your to be had assets, and favored intention. For example, do you want to see if a targeted intervention relieves emotions of anxiousness? Essentially the most convincing results for that might come from a true experimental design with random sampling and random mission to groups. Eventually, this can be a selection that will have to be made in close collaboration with your consultant. Hence, I recommend discussing the professionals and cons of each sort of study, what it would imply for your private dissertation method, and what's required of each and every design before making a choice.
Experimental research designs are centered on a transparent speculation, the rationale of the research is to confirm or refute the validity of the speculation. Experimental study designs have an unbiased variable, a stylish variable, and a manage staff. Lots of the experiments are performed in a laboratory in a managed environment. The reason of the test is to discover the causation and experimental reports are causal reviews. These stories answer what, why and even how questions within the study. The experimenter can manipulate the variables and he has a manage workforce and a placebo. The manage workforce receives the medication that the experimenter needs to test and the placebo staff is tested with none healing. The change in the outcome of both businesses are when compared. The experimenter repeats the scan in the equal atmosphere multiple time to get most valid outcome. There are basically three unique forms of experiments: managed experiments, quasi-experiments, and subject experiments.
Methodology
The experimenter determines two corporations to perform an test: one workforce is known as the experimental crew and the opposite group is known as a placebo team. The experimental crew will get the healing and the placebo does not get any treatment. The experimental staff is much like the control staff besides that the manage staff does now not get any medication at the same time the experimental team gets the healing. The experimenter repeats the experiment twice or thrice to develop the validity of the results.
Non-experimental research design
Non-experimental research designs are carried out in normal settings, it does now not involve manipulation of the crisis, occasion, instances or persons. Survey, case reviews, correlational experiences comparative stories and descriptive experiences are one of the vital examples of non-experimental study design. Longitudinal studies are also non-experimental research and the reason of those experiences is to study a obstacle, persons or phenomenon over a period of time to detect the exchange.
A correlation will also be founded utilizing non-experimental research design however causation are not able to be established. To establish causation, the researcher will have to be ready to assert that the result is the outcome of the found variable and now not whatever else. They don't have a control staff and the study design is incredibly bendy. Due to the absence of the manipulate crew the researcher are not able to ascertain that the final results are the direct outcome of the variable that has been studied. The non-experimental study design study the phenomenon, folks or trouble in a natural surroundings without manipulating it, accordingly, the findings can be applied to a extensive audience.
Methodology
In a non-experimental study, the researcher does no longer manipulate the variables to be confirmed and as a consequence he can't ascertain the effect of the independent variable on the elegant variable. A causation is established in one of the vital nonexperimental reviews but no longer in all of them. The causation is established via picking that a detailed variable has an have an impact on on another variable. It can also compare the outcome of two or extra businesses or individuals on one or more variables.