Desktop version

Home arrow Sociology arrow The future of mobility

Integrate into Scenario Frameworks

The research team used two distinct types of analysis-cross-impact analysis and consistency analysis—to identify the values that would group descriptors and projections into distinct scenario frameworks (for a more-detailed discussion of these analytical tools and how they compare with other analytic methodologies, see Amer, Daim, and Jetter, 2013). We first used cross-impact analysis to describe the relationships among the descriptors. The team developed a cross-impact matrix that matched each pair of descriptors across all influencing areas. The team determined whether it was plausible that either of two descriptors affected the other, using a four-point scale in which 0 indicated no influence and 3 indicated a strong influence. For example, we determined that the total population has a strong influence on urbanization, but not the other way around. In our analysis, we did not carry forward descriptor pairs that we determined to be totally independent; however, they did remain part of the final scenario frameworks. This assessment also forms the basis of analysis into how active and passive each descriptor is (see Figure A.2 in Appendix A). Active descriptors tend to be drivers of the scenario, in that changes in them are thought to produce different outcomes between scenarios.

Second, we applied consistency analysis to those pairs of descriptors in which one was found to influence the other. This analysis examined the various projections for each descriptor. RAND and ifmo staff jointly developed this consistency matrix. Each pair of projections for the two descriptors was rated on a five-point scale, from totally inconsistent (1) to strongly consistent (5). For example, strong growth in demand for air travel was deemed consistent with the lower price of oil because demand is affected by price, and higher oil prices mean more-expensive airfares. On the other hand, because the eastern region is already wealthier than the rest of the country and greater concentration of economic activity in one region is likely to lead to a concentration of wealth as well, we deemed an increasing concentration of economic output in that region to be inconsistent with a decrease in income inequality.

Then, ifmo fed these results into an online tool called the RAHS platform to group specific projections across all influencing areas. Of all the mathematically possible groupings of projections, RAHS eliminated those that contained total inconsistencies (as defined in the consistency matrix). Of those remaining, RAHS identified clusters of descriptors and projections that formed four unique and complementary scenario frameworks.

Although there is not a hard and fast limit on the number of scenarios that should be developed, selecting those that differ most meaningfully requires expert judgment. Of the four clusters produced, the research team selected two to develop further. Three of the clusters were similar across many descriptors, so we decided that only one of the three should be carried forward. We included the fourth cluster specifically because it differed so greatly from the other three. For example, it had a lower average gross domestic product (GDP) growth rate. We thought it was important to include different growth rates because, at China's stage of development, economic growth is generally an important, though hardly the only, determinant of mobility. From the group of three clusters, we selected the one that differed most from the lower-GDP-growth-rate cluster because of its different projections for total population, share of the economy in the eastern region, and domestic vehicle production. These two frameworks became the basis for the scenario narratives.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics