When are Statistical techniques not applicable?

L

leomoon45

I just came from a Lead Assessor Training here in the Philippines. My trainer there tells us that don't say the clause on Statistical technique is not applicable. Any evidence that shows the company is doing what it can to analyse data should be part of this element. But our Australian consultant tells us Statistics yes, but Statistical Analysis is an altogether different thing, ergo, NA. I've had at least 2 assessors saying no, and 1 says yes. Your site offers advice to not include SPC if its not applicable.

We're in the convenience store industry.


________________________
ahsan
 

Marc

Fully vaccinated are you?
Leader
Re: When are Statistical techniques not appplicable?

Your Australian consultant is correct. SPC is one kind of "statistical technique". Analysis of data is required, but not necessarily SPC.
 

Steve Prevette

Deming Disciple
Leader
Super Moderator
Re: When are Statistical techniques not appplicable?

Over 23 years of dealing with statistical techniques, I always run into the folks who say "we'll do SPC when it's applicable" which I've learned translates as "No way, no how, are we doing SPC".

As far as when is it truly not applicable, I would say in very short term projects (less than a year) with specific budget, schedule goals - did I meet the goal or not, it may not be "Applicable". Just running the punchlist may be sufficient. But if you are a company that does a series of short term projects, you may want to use SPC (or other statistical analysis) in order to study how well you do at setting and meeting goals over the long run.

I will suggest (as a degreed Operations Researcher) that of the various tools out there called "statistical analysis", SPC is the simplest and most intuitive to non-statisticians. Note, I don't consider moving averages to be "statistical analysis".

See http://www.efcog.org/wg/esh_es/Statistical_Process_Control/
 

Statistical Steven

Statistician
Leader
Super Moderator
Re: When are Statistical techniques not appplicable?

I must disagree with my colleagues here. Statistics is a science like Chemistry. Using the knowledge of statistics and APPLYING them to the process is statistical techniques. So from an audit perspective one can say that calculating means and standard deviations are statistical techniques. The intention is not to BLINDLY apply statistical techniques but to make INFERENCES from the data that are actionable. So my take is that applying statistics, statistical techniques and statistical analysis without any inference is not following the clause.
 

Marc

Fully vaccinated are you?
Leader
Re: When are Statistical techniques not appplicable?

I guess a question, to help the OP a bit further, would be what statistical techniques might be appropriate for a convenience store.
 

Steve Prevette

Deming Disciple
Leader
Super Moderator
Re: When are Statistical techniques not appplicable?

SPC. Hmm answer is too short need to add some characters.
 

Marc

Fully vaccinated are you?
Leader
Re: When are Statistical techniques not appplicable?

Let's forget SPC in this case. I doubt there is much if anything SPC would be applied to in a convenience store. I may be wrong, of course.

I doubt there are many convenience store people here to help. I was just trying to think of areas where data analysis might be appropriate to use.
 

Steve Prevette

Deming Disciple
Leader
Super Moderator
Re: When are Statistical techniques not appplicable?

We do then need to step back to - what are the metrics that a convenience store could keep?

I would assume that includes the usual budget issues - purchasing costs, sales revenue. Then you may have issues like loss from spoilage, loss from threft. Employee turnover, perhaps employee suggestions and morale surveys (if they exist). Customer satisfaction surveys, if they exist.

I'll even volunteer that if whomever is asking the question can send me on the order of two years of monthly data to my home email or PM here, I'll work them up an initial set of control charts.
 

Bev D

Heretical Statistician
Leader
Super Moderator
Re: When are Statistical techniques not appplicable?

I guess it depends on what is meant by the convenience store industry. Is the OP a supplier of things that convenience stores use or does his/her organization run a chain of convenience stores, or?

one can use statistical modeling for understanding sales volumes in order to plan staffing and material buys.

types of DoE and statistical analysis can be used to understand and improve the effectiveness of marketing, sales promotions and material placement.

If the OP is involved in the manufacture and/or sale of things then the might be interested in the reliability of the equipment, its ease of use or attractiveness to promote sales.


I agree with Statistical Steven that the standard is not prescriptive about what statistical techniques are used, but that if 'numerical' analysis is performed then appropriate statistical methods should be applied so that the inferences and conclusions are properly supported. the lessons that the red bead experiment and the funnel experiment teach us are exactly why statistical methods are preferable over 'gut feel'.

SPC is almost always appropriate whenever looking at time series data, rather than 'eye-balling it' or worse using the pseudo statistical 'trend line' from EXCEL.
 
Top Bottom