[MUSIC] After having this class, the indicators of the KOF Youth Labor Market Index, what is the first step to aggregate the indicators into a single index? >> Before aggregating the indicators into a single index, we rescale each indicator value into an indicator score that takes values between one and seven. We do so based on these formulas. This has three reasons. First, while most indicators are expressed as rates that vary between 0 and 100, most of the indicators also take values within a relatively narrow range that differ substantially across indicators. For example, while the unemployment rate has a mean of 18%, the mean of the formal education and training rate is 60%. >> Thomas, I have another question. Can you show us what happens if we don't rescale the variables? >> Yes, this spider graph shows the results for Lithuania, Greece, the EU 28 and developing countries, using the original variables that range from 0 to 100. As you can see, this graph is very hard to read, because differences across countries remain invisible. The spider graph for indicator values between 1 and 7, on the other hand, allows to visualize the differences. >> Thomas, you mentioned two other reasons to rescale the variables, what are those? >> The second motive for rescaling indicators arises because for some indicators, in our case the formal education and training rate, higher values indicate an improvement, while in other cases higher values indicate a deterioration. Rescaling harmonizes the direction of the indicators. Third, rescaling indicators allows us to compare indicators that have different scales. In our case, this would be the scales mismatch rate and the relative unemployment ratio. >> We are all the time talking about an index. How are rescaled indicator values aggregated into a single index? >> The last step towards calculating the aggregated index consists of defining weights for each indicator and dimension. We take a very simple approach to choosing the weights of the dimensions and indicators. First, we give equal weight to the four dimensions, so that each dimension has a weight of 25% in the KOF Youth Labor Market Index. Second, we give equal weight to the indicators in each dimension. By doing so, the indicators in the activity state dimension each have a weight of 8.3%. The working conditions indicators each have a weight of 5%. And the indicators in both the education and the transition smoothness dimensions each have a weight of 12.5%. Now we are ready to go back to the online tool and discuss the displayed results. After clicking next, the first graph we see is the spider graph. We have discussed this already, but can you explain to us what the scoreboard illustrates? >> If you click on Scoreboard, you can see the values, scores and ranks of each indicator shown in the spider. Furthermore, the percentage weight column shows the weight of each indicator and dimension carried in the baseline index. >> Can the user change these weights? >> Yes, they can. This is an important function to allow, because the relevance of indicators might differ across countries. For example, the indicators in work at risk of poverty rate and the vulnerable employment rate might be more important in emerging economies than in developed countries. Therefore, the KOF Youth Labor Market Index tool allows the user to choose his or her own weighting schemes in the weights column. Here is an example of changing an indicator's weight to zero, in this case for the relaxed unemployment rate. This doesn't affect the spider graph, which shows the multidimensional nature of the youth labor market situation. However, as you can see, it changes the aggregate index score, which is shown in the last column of the Scoreboard. This option is particularly useful in cases where the data is incomplete for a country. Choosing a weight of zero allows you to calculate the KOF Youth Labor Market Index based on a comparable set of indicator. In the present example, we would choose a weight of zero for the indicators relaxed unemployment rate and temporary worker rate, since these indicators are missing in Lithuania. >> What is the meaning of the exclamation mark in the top right corner? >> The exclamation mark indicates that there are a lot of missing values. These might arise because either of many indicators are missing for a few countries or a few indicator values were missing for many countries. Consequently, this issue of missing data needs to be even more carefully considered for country groups than for individual countries. By clicking on index over time, you get the graph in the upper part that displays the development of the use youth labor market situation over time. >> What is the meaning of the bars in the lower part of the graph? >> These bars show the number of available indicators. As you can see on the left side, only relatively few indicators exist in the 1990s, particularly for Lithuania. This suggests that the aggregated KOF Youth Labor Market Index is not fully comparable in this time period. In 2000, the number of indicators jumped up for all four series. While the KOF Youth Labor Market Index remains relatively stable for the EU 28 and developing countries, it increases substantially for Greece and drops down for Lithuania. This highlights that the availability of data for the indicators may play an important role in the interpretation of the aggregated index. [MUSIC]
