Read the summary and the most important questions on Empirical Research Project

• Lecture 1: Measurement

  This is a preview. There are 7 more flashcards available for chapter 02/02/2016
  Show more cards here

• What are the different types of variables that can be conducted during a research?

  1. Categorical variables, binary (yes/no), nominal (e.g. colors, jobs etc.) and ordinal (e.g. first, second and third price). Continuous variables, interval (e.g. of a scale from 1 -5).Ratio (same as interval only with absolute zero e.g. length). Discrete (only with certain values e.g. whole numbers).

• Why is primary measures sometimes more reliable than secondary data, according to Hult et al. (2008)?

  The use of primary measures for measuring performance in IB is particularly appropriate when the researcher is aiming to identify not only the goals associated with a specific strategy, but also the understanding and interpretation of an organization's performance goals by managers. In situations where firms (and managers) are hesitant to provide secondary financial, operational, or overall effectiveness performance data, collecting primary data provides IB researches with a better ability to understand the values that a manager may place on specific financial, operational, or overall effectiveness performance measures.

• What has Hult et al. (2008) concluded about cross-sectional and longitudinal data?

  They state that cross-sectional measurement of performance is often insufficient to support the casual inferences that researchers desire to make in performance studies Ideally, performance and its antecedents should be measured longitudinally. By gathering data across time periods.

• What does Hult et al. (2008) say about endogeneity and selection?

  They argue that although IB research may be limited by data availability, statistical tools can often be used to address complications related to endogeneity and selection. As a value-added element, IB researchers should consider and report issues of endogeneity and selection in their studies, to allow future research the possibility to more fully explore such issues.

• What types of data sources are there?

  Primary data
  

  • Collected by the researcher
  • Subjective (the ‘objectiveness’ and ‘subjectiveness’ is about how the data is gathered)
  • Raw material
  • Personal prejudice (selection of the data)
  • Expensive/time-consuming
  • More original
  • Usually suitable for the analysis

  
  Secondary data
  

  • Collected by others
  • “Objective” (the ‘objectiveness’ and ‘subjectiveness’ is about how the data is gathered)
  • Ready made products 
  • Possibility of lesser degree personal prejudice (selection of the data)
  • Cheaper
  • Lacks newness
  • May or may not be suited for the requirement of the analysis

• What different level of analysis are there?

  
  
  
  
  1. Firm
  2. Strategic business unit(SBU)  
  3. Inter-organizationa lunit 

• Hult et al. (2008) describe how a focus on firms or countries with comparable accounting standards may cause a selection bias. Give at least three other examples of how selection bias may occur in studies of performance.

  • Convenience bias (availability of data)
  • Popularity bias (most popular measures)
  • Resource bias (least costly measure)
  • Data source bias (overlooking measures)

• In an IB research project on performance, how to determine which type(s) of performance measures to include?

  • Depends on the research question and the ‘problem’ you want to solve, by identifying the measures (e.g. financial, operational or other) that match the hypotheses that needs to be tested
  • Data source influences the type of performance measure (e.g. the limited amount of time to interpret data)
  • Level of analysis influences which performance measure is most suitable (e.g. worker/team/department/business unit/MNE)
  • Internal validity
  • External validity

• Lecture 2: Descriptive statistics

  This is a preview. There are 21 more flashcards available for chapter 09/02/2016
  Show more cards here

• What are the first two different measures of dispersion?

  Skewness: 
  - Symmetry of the distribution
  - Kind of skew labelled according to the longer tail
  - Skewed distributions = skewness values
  
  Kurtosis:
  - degree of which socres cluster in the tails of a frequency distribution
  - peakedness/flatness relative to the normal distribution.
  - Leptokurtic (high peak) Mesokurtic (middle peak) Platykurtic (flat peak)

• What are the 'third and fourth' measures of variability or dispersion?

  Range: Maximum value - minimum value
  Quartiles, Interquartile range.
  Variance or standard deviation