The help text for statistics does not correctly and unambiguously describe how the trends for the 7-Day Incidence and the 7-Day R Value are calculated.

The information text for "Key Figures" says "The arrow direction indicates whether the trend is increasing, decreasing, or remaining steady – that is, demonstrates a deviation of less than 1% compared to the previous day or 5% compared to the previous week. The color indicates this trend as positive (green), negative (red), or neutral (gray). The trend compares the value from the previous day with the value from two days ago or, for the 7-day trends, the average value from the last 7 days with the average value from the 7 days prior to that."
Each of the values showing a trend has a label containing the text "7-Day" so it seems that all the trends should be calculated using the value from one week previously with a 5% band.
appDownloads7daysTrend5percent in StatisticsJsonStringObject.java#L19.sevenDayRvaluePublishedTrend1percent in StatisticsJsonStringObject.java#L73. Also confirmed in https://github.com/corona-warn-app/cwa-documentation/issues/535#issuecomment-799024445.The information text for "Key Figures" should correctly describe how the trend is calculated for all values. For 7-Day Incidence and 7-Day R Value the text should make it clear that the trend calculation is based on comparing the value from the previous day with the value from two days ago using a 1% steady band.
Comparison
Value \ Steady band % | Daily
1% | Weekly
5%
--- | --- | ---
Confirmed New Infections
7-Day Average | | ✔️
Warnings by App Users
7-Day Average | | ✔️
7-Day Incidence | ✔️|
7-Day R Value | ✔️|
Change the description in the string statistics_explanation_trend_text from Corona-Warn-App/src/main/res/values/strings.xml in Android and in the equivalent string in iOS.
It should be clear that the 7-Day Incidence and for the 7-Day R Value trend calculations use the comparison between the previous day and two days ago.
For Android, change the string statistics_explanation_trend_text:
from
"The trend compares the value from the previous day with the value from two days ago or, for the 7-day trends, the average value from the last 7 days with the average value from the 7 days prior to that."
to
"The trend compares the value from the previous day with the value from two days ago or, for the values labelled '7-day average' , the average value from the last 7 days with the average value from the 7 days prior to that."
from
"Der Trend vergleicht den Wert vom Vortag mit dem Wert von vor zwei Tagen bzw. für die 7-Tage-Trends den Mittelwert der letzten 7 Tage mit dem der vorausgegangenen 7 Tage."
to
"Der Trend vergleicht den Wert vom Vortag mit dem Wert von vor zwei Tagen bzw. für die Werte, die als 7-Tage-Mittelwert bezeichnet sind, der Wert der letzten 7 Tage mit dem der vorausgegangenen 7 Tage."
For iOS a similar change should be done.
See also FAQ: Trend differences between 7-Day Average and 7-Day Incidence for a suggestion about modifying the FAQ entry https://www.coronawarn.app/en/faq/#further_details. This only deals with the 7-Day Incidence issue, not with the question about the 7-Day R Value.
Internal Tracking ID: EXPOSUREAPP-5225 from https://github.com/corona-warn-app/cwa-documentation/issues/528
@GisoSchroederSAP confirmed in https://github.com/corona-warn-app/cwa-documentation/issues/535#issuecomment-799024445 about the 7-Day R-Value using the previous day as a comparison.
I have updated the original post to now include concrete text for a possible fix.
Where the help text in English refers to the "7-day trends" it does not mean all values labelled "7-Day", it means just the ones labelled "7-Day Average". The other two values, "7-Day Incidence" and "7-Day R Value" are also 7-day trends, but the trends for these two values are calculated based on comparison to the previous day, not to the previous week.
The help text problem can be solved by making the text more precise and by explicitly referring to the values according to their labels "7-Day Average". Only for the values with a label "7-Day Average" are the trends calculated by comparing to one week previously.
Hi @MikeMcC399 ,
just for the records: Yes, we now have confirmation an deeper insight. Yes, the text on the information screen in the App need to be adjusted, we all agree.
However, one more "wording" in your last post: Let's please differentiate the _value_ from the _trend indicator_ (the arrow with color): None of the "7-Day Incidence" and the "7-Day R value" by themselves are trends, they are _values_, just calculated on a given day for a 7-day period.. The _trend_ for both metrics appears only out of the comparison to the values with the ones from the previous day, respectiveley.
So, I assume, we are back to the "naming convention" - if someone reads "7-day ???" on different metrics this often is misleading to the understanding all those numbers reflect (the same) trends.
Anyway, we are working hard to document the FAQ and to adjust the App text blocks. Thanks for your continuous contribution here, this is much appreciated!!!
Hi @GisoSchroederSAP
You are right of course that the values and the trends of the values are different entities. The values of the trends are not displayed in numerical terms only in color and arrow direction.
So, I assume, we are back to the "naming convention" - if someone reads "7-day ???" on different metrics this often is misleading to the understanding all those numbers reflect (the same) trends.
That is exactly the problem, that the help text is open to wrong interpretation. The help text needs to be written so that there is a high likelihood that any normal reader will interpret it correctly.
Anyway, we are working hard to document the FAQ and to adjust the App text blocks.
Good to hear! Thank you!
@GisoSchroederSAP
I updated the original submission to make it clearer about values and trends of values.
@GisoSchroederSAP
Anyway, we are working hard to document the FAQ and to adjust the App text blocks.
I can't see any change yet. Would you like to give an update on your progress? Although there have been no more detailed questions from users on this topic here in GitHub it would still be good to clarify the documentation.