Cross impact analysis is a method used in forecasting exercises aimed at measuring the correlation between future events (variables).
Mainly in the field of technological developments, CIA is applied to identify how developments in one area interact with those in another, how strong the mutual influence is and in how far it affects the outcome of others. CIA’s main asset is its ability to show how one situation impacts another situation.
CIA is commonly based on expert opinion, exploring the likelihood of an event or trend and its dependency on the occurrence of other developments or impacts. Thereby, points of agreement or divergence between experts can be identified. It is often thought of as an extension of the Delphi Survey and frequently used in scenario analyses (Simon & Motavelle 2006).
Applying the method
Cross impact analysis involves constructing a matrix to show the interdependencies of different events and basically includes the following steps:
- Identification of events and trends to be considered (usually between 10-30) and experts to be questioned
- Estimating the probability for each event (in isolation of the others)
- Calculating the probability of each event in dependence of the others
- Sensitivity analysis
- Generation of scenarios and the CIA matrix.
The matrix lists the set of events or trends that may occur along the rows, and the events or trends that would possibly be affected by the row events along the columns. Respondents are then required to assess how the occurrences in each of the rows affect the probability of the event in the corresponding column. The person analysing the results can average the responses to generate a summary; this summary is known as a cross impact analysis (Simon & Motavelle 2006)
According to the JRC/IPTS carrying out a CIA (steps 1-5) takes a minimum of 2 to 8 months though can vary in case further tools or methods are applied (JRC/IPTS 2006).
Strenths & Weaknesses
+ considers the interrelation of events
+ reveals differing perspectives on future developments
- selection of experts and events is crucial for a CIA’s outcomes
- is based on pairs of data, however, several events can influence another one at the same time
- can be time and resource consuming, depending on the number of events taken into account
JRC (Joint Research Centre), IPTS (Insitute for Prosepctive Technological Studies) (2006). For-Learn. Cross-Impact Analysis. http://forlearn.jrc.ec.europa.eu/guide/2_design/meth_cross-impact-analysis.htm#Steps (19 Dec 2013).
Simon, K.-H., Motavelle, A. (2006). Cross impact analysis. SustainabilityA-Test.
Wikipedia Cross-impact analysis.