A confidence interval is an interval for a parameter of a statistical model. A hypothesis that states that the parameter has a certain value can be rejected if the value is not in the confidence interval. If the value is in the confidence interval the hypothesis cannot be rejected. In this sense a confidence interval is an in terval of acceptable hypotheses. Using 93 % confidence intervals means that 93 % of the times a confidence interval is calculated it will contain the true value of the parameter. Usually one uses confidence one levels of 90 %, 95 %, or 99 % and each discipline has (or should have) its own standards.

My suspicion is that the use of a 93 % confidence interval is an adaptation to data rather than a predetermined standard. If my suspicion is right it means that the standard of what is considered as acceptable has been changed in order to reach a certain conclusion that the scientist prefer.

Hi,

The confidence interval (also called margin of error) is the plus-or-minus figure usually reported in newspaper or television opinion poll results. For example, if you use a confidence interval of 4 and 47% percent of your sample picks an answer you can be "sure" that if you had asked the question of the entire relevant population between 43% (47-4) and 51% (47+4) would have picked that answer.

The confidence level tells you how sure you can be. It is expressed as a percentage and represents how often the true percentage of the population who would pick an answer lies within the confidence interval. The 95% confidence level means you can be 95% certain; the 99% confidence level means you can be 99% certain. Most researchers use the 95% confidence level.

For more info see please http://www.surveysystem.com/sscalc.htm

Hope this helps.

Why use 93% ?

Without knowing details I would assume that 95% or 99% CI would have resulted in much lower numbers, thus undermining the case for assuming a relationship between A and B.

Kind Regards

Roland

A confidence interval is an interval for a parameter of a statistical model. A hypothesis that states that the parameter has a certain value can be rejected if the value is not in the confidence interval. If the value is in the confidence interval the hypothesis cannot be rejected. In this sense a confidence interval is an in terval of acceptable hypotheses. Using 93 % confidence intervals means that 93 % of the times a confidence interval is calculated it will contain the true value of the parameter. Usually one uses confidence one levels of 90 %, 95 %, or 99 % and each discipline has (or should have) its own standards.

My suspicion is that the use of a 93 % confidence interval is an adaptation to data rather than a predetermined standard. If my suspicion is right it means that the standard of what is considered as acceptable has been changed in order to reach a certain conclusion that the scientist prefer.

It may be that the researcher wanted to have the observed value of the statistic IN the confidence interval.

AND, the length of the confidence interval reflects precision of the estimate. To choose a smaller confidence level is accompanied by shorter intervals reflecting larger precision.

Thus, to choose the smallest level of confidence that still contains the observed value of the statistic leads to a better precision of the estimation related to this interval.

Methodologically however,  it is required to select the level of confidence BEFORE the data are collected.

Without details no final interpretation can be stated. Let me point, however, at the possibilty, that  "93% confidence" is meant as equivalent of "p-level equals 7%".  Obviously  it this case the formulation should be also much more precise.

Regards

As p-levels are a bit difficult to understand - at least if one knows already that tthey do not provide a probability for the hypothesis tested - there might come up a new trend to report the results with the smallest possible confidence interval (if an interval can be constructed) and report its confidence level.

@Manfred Borovcnik I guess that what you call "the observed value" is the same as the maximum likelihood estimate of the unknown parameter. If so the "observed value" is always in the confidence interval. The question is whether a value of the parameter corresponding to a nul-hypothesis is in the interval. For any nul-hypothesis one can select a confidence level such that the nul-hypothesis is accepted. A 93% confidence level correspond to significance level of 7 %. As mentioned by Joachim Domstra the notion of p-value and significance level may have been mixed up. As you point out significance level has to be selected before data are collected.