EMERGENCY HELP REQUIRED (within 12 hours)

[† please_help_me †]

Im sure you all know what error bars are, in application to a graph. Basically, i have a grapgh with error bars on the points. I also have a line of best fit. Unfortunately this line of best fit doesnt lie on the margins of the error bars for 1 point out of 9. What does this mean? That it is an anomalous value? That the error bars need to be increased?

A swift reply would be GREATLY appreciated. Thanks in advance

[Astronuc]

It is quite possible to have one or more data points with error bars that do not intersect with best fit curve. It hard to know the situation without knowing the data, i.e. indepedent and dependent variables.

The error bars reflect experimental uncertainty perhaps, and it is possible to have high precision/accuracy yield small error bars, and therefore have an 'outlier' in which one of the error bars does not intersect with the best-fit curve. If the outlying datapoint is at the end, then perhaps a different function (curve) would be appropriate. If the other datapoints fall tightly above and below the best fit curve then, it is likely an appropriate best fit.

A single outlier could be an anomalous point, i.e. perhaps there is a measurement error or perhaps another variable/factor affects the results.