This workshop will provide you with an up-to-date overview of basics, techniques, methods, tricks, and tools for Internet-based experimentation. Using a practical example, you will learn how to conduct Web experiments that provide results within hours. For this purpose, the workshop includes a hands-on demonstration of interactive systems for designing Web experiments online, e. You will learn about ways of dealing with issues in experimental design, security, recruitment, sampling, self-selection, multiple submissions, question design, dropout, error estimation, distributed experimentation, data handling, data quality, and log file analysis.
Among other "tricks", you will know the high hurdle technique, the warm-up technique, sub-sampling procedures, multiple site entry, and ways to check for motivational confounding. You will learn how to make use of Web services like the free "::web experiment list::", and of analysis tools for subsequent data analysis, such as Scientific LogAnalyzer. Learn or review basic knowledge about Internet-based experimenting; learn a number of useful techniques and tricks; learn to avoid pitfalls that may make your experiment useless; create an experiment during the workshop; get to know several tools and how to use them.
Reips, U. Standards for Internet-based experimenting. Experimental Psychology, 49 4 , PD Dr.
Contact Report 260
As a methodologist he has conducted research to the advancement of methodologies and techniques for Internet Science, especially on Internet-based experimenting. He has given workshops on this topic in several European countries as well as in the US e.
Anhand eines praktischen Beispiels werden wir ein Web-Experiment aufsetzen. Personen die bereits erste Erfahrungen mit Online-Befragungen gesammelt haben. Batinic jku. Im Rahmen des Workshops werden verschiedene erfolgreiche und weniger erfolgreiche Praxisprojekte aus der Online-Marktforschung vorgestellt. Although many web-based studies have investigated scale types using radio buttons and numeric box entry, very few studies have examined the possibility of use of visual scales like a Visual Analog Scale VAS or Graphic Rating Scale GRS.
Respondents rated the importance of a set of 5 items either political priorities or features of a grocery store. Immediately following, they rated their impressions concerning the scales difficulty, accuracy, interest. Using a standard grid format, they were then asked to rate how much each feature would affect their intention political issues: likelihood to vote for a spending increase for the issue; grocery store: likelihood to shop at a store with the feature.
Banking all 5 items on the same screen took about We will also present some distributional differences that we observed between the scale types. In general, the 5 segment GRS had lower means, whether banked or not. SniperScale: Graphical scaling in data collection and its effect on the response behaviour of participants in on-line studies. Using graphical support, questionnaires can be more interesting for the participant and motivate to complete the entire interview.
In a study with more than 2, respondents, an experiment testing the methods was integrated for the examination of the possible effects of the applied graphical scaling. The study used three technologically different approaches of scaling which were randomly integrated. Each case represented an identical Analytic Hierachy Process scale in a different appearance:. Classical representation on the basis of radio buttons with graphical support. Application of a graphical slider based on Java technology with graphical support. The results achieved lead to an improvement of the creation of scales in the future:.
The SniperScale seems to be an interesting way to entertain the respondent while interviewing him without losing quality. The study shows clearly that there is a trade off between quality and costs and the appropriate treatment of the respondents when deciding about graphical scaling support. Randall K. Thomas , Jonathan D. Klein, Susan Behnke, George Terhanian.
Harris Interactive, USA University of Rochester, USA Harris Interactive, UK. In this current study, we examined how response formats could affect behavioral intention measurement. To assess respondent satisficing, 2 manipulations varied respondent load. Respondents were randomly assigned to complete the experimental section either at the beginning of the survey or at the end of the survey and were randomly assigned to either 5 or 10 behaviors out of 20 behaviors e.
- The Wealth Code: How the Rich Stay Rich in Good Times and Bad.!
- Columbiahalle - Veranstaltungen - deutsch;
- Die Dialektik der historischen Aufklärung gegen die AfD.
- The Future Of Mankind - A Billy Meier Wiki - Contact Report !
As a standard criterion, respondents were asked how frequently they engaged in the assigned behaviors in the past 30 days and then randomly assigned 1 out of 6 possible likelihood response formats asking if likely to perform the action in the next 30 days :. Replicating prior work, the YNG response format yielded significantly higher likelihood endorsement frequencies across behaviors than the MRL, regardless of survey position.
Extending our findings, the dichotomous likelihood measures also had higher endorsement across behaviors than the MRL. Behaviors presented later in the survey had somewhat lower rates of endorsement as being likely, especially for the MRL, though this effect was observed for all formats. Behaviors presented in a context of 10 behaviors showed somewhat lower likelihood endorsement than the 5 behavior context. In terms of criterion-related validity, the YNG format explained somewhat less behavioral variance than did the MRL format. For the likelihood scales, the 5 category unipolar scale predicted substantially more variance than any of the other scale types, replicating earlier research on behavioral intention measurement Thomas, Behnke, and Johnson, , AAPOR; Thomas, Terhanian, and Bayer, , RC It's a convention within the social science to handle 5-point or 7-point scales as if they are interval scales.
Indeed, everybody knows that this is wrong, but the convention is necessary for the usage of statistical analysis. This idea is not new; it's due to the fundamental research of the US-psychologist S. MES never established within market research though this method is more selective than others. The authors carried out an online employee survey at Lufthansa Technik by using MES for the test about the intranet presence of the Marketing and Sales Department.
They present the results in comparison to a test design using a 7-point scale. Interested person receive a handout of the presentation at the topcom booth MML Mediencampus, Foyer 2 nd floor and could see a live demonstration of the method as well. Es ist eine sozialwissenschaftliche Konvention 5er- oder 7er Skalen wie Intervallskalen zu handhaben. Die Ergebnisse werden im Vergleich zu einem Testdesign mit 7er-Skala vorgestellt. The importance of usability for the success of a webshop has been shown in many empirical studies.
For the identification of usability problems, the methods of thinking-aloud, video-cued-thought-protocols and standardized-questionnaires are often used. Each method requires a different expenditure of time and causes different expenses.
libretext/guangconbearsnow.cf at master · garethsprice/libretext · GitHub
Hence, the selection of a useful method is of great importance. Therefore, the following questions should be answered: How many and which usability problems can be explored by each method? Which are the specific pros and cons of each method concerning the task performance, the disruption of the ongoing surf-process, the awareness and evaluation of the test situation?
To answer these questions, three different methods thinking-aloud, video-cued-thought-protocols and standardized-questionnaires — were tested in an experimental study 65 participants; 10 tasks; real webshop. Each method combined the verbalization thinking-aloud, video-cued-thought-protocols or a questioning and an observation based on screen-records. For the identification, we could distinguish whether a problem was discovered sorely by the verbalization or observation alone or if the verbalization as well as the observation identified a usability-problem.
All other variables were measured by a questionnaire. Based on our findings, we can evaluate the usefulness of each method. Regarding the amount of usability-problems that were discovered through the combined use of a verbalization-technique and the observation, the thinking-aloud-method came up with a better result compared to the method of video-cued-thought-protocols. Not to use one of the two verbalization-techniques appeared to be unfavourable, since the combination of an interview and the observation could only identify fewer problems compared to the other methods.
Concerning the task performance time, clicks, results , no significant difference could be found between the three methods. In contrast to this objective finding the subjective experience of the participants was contrary. Furthermore, participants evaluated the test-situation to be more realistic in the case of the interview. In a comparison of the two verbalization methods no significant differences could be found regarding the subjective estimation of the expenditure of time for the task performance and the evaluation of the test situation. The results show that in addition to expenditure of time and costs many important factors should be addressed by the selection of a useful method.
In zahlreichen Studien hat sich gezeigt, dass die Benutzerfreundlichkeit bzw. Jede Methode erfordert bei ihrem Einsatz einen unterschiedlichen Zeitaufwand und verursacht unterschiedliche Kosten. Daher stellen sich bei Auswahl der Methode n die Fragen: Wie viele und welche Usabilityprobleme deckt sie auf?
Alle weiteren Variablen wurden mit Hilfe eines Fragebogens erhoben.
Hinsichtlich der Task Performance Zeit, Klickanzahl, Aufgabenergebnis lassen sich keine signifikanten Unterschiede bei den drei Methoden feststellen. Beim subjektiven Eindruck der Probanden zeigt sich jedoch genau das gegenteilige Ergebnis. Traditional scale development usually draws on reflective measurement models in which the observed indicators are assumed to be caused by the latent variable.
In contrast, in case of formative constructs it is postulated that the indicators cause the latent variable. High correlations between formative indicators might occur, but are not generally expected. Hence, factor analysis and Cronbachs Alpha are inappropriate methods of scale evaluation. It has been shown that measurement models have been misspecified as being reflective in many studies, although the constructs afford formative models. In consequence of this kind of misspecification reduced scale validity might result.
Both one formative and one reflective usability scale were developed and administered to subjects who participated in a laboratory experiment for evaluation purposes, each exploring two of a total of 35 online stores. Among usability and other predictor variables, the intention to buy and actual purchase behaviour were assessed as dependent variables.