This thesis deals with the conjoint analysis in the context of the possibilities for its use in social science research. The mentioned method is introduced in detail and the text covers both the historical development and statistical background of each type of this analysis. Following practical part of thesis deals with the ability to overcome some typical data validity issues such as social desirability and order effect. Hypotheses, formulated on the basis of literature, are being tested using data obtained from split ballot experiment by comparing outcomes of conjoint analysis with those obtained with conventional rating scales. As the main topic of the conducted study is measuring preferences within individual aspects of ideal job, analysis no surprisingly identifies the salary as the most important attribute, but besides this, conjoint analysis shows potential for being more resistant to the order effect than the standard rating scales.
Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:332647 |
Date | January 2015 |
Creators | Sklenařík, Pavel |
Contributors | Soukup, Petr, Jeřábek, Hynek |
Source Sets | Czech ETDs |
Language | Czech |
Detected Language | English |
Type | info:eu-repo/semantics/masterThesis |
Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0018 seconds