LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (April 2006)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Wed, 19 Apr 2006 06:53:37 -0500
Reply-To:     "Beadle, ViAnn" <>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Beadle, ViAnn" <>
Subject:      Re: effects of road diversions on shops
Comments: To: Omar Farook <>
Content-Type: text/plain; charset="iso-8859-1"

What you should do is just report what you said below and provide summary descriptive statistics such as xx% of shop owners report that they have been negatively affected. It might also be useful to merge in information about the timing and extent of road diversions. If every shop in the study has been subject to road diversions, you really can't do much more analysis. I personally would want to also take a closer look at the respondents who reported no negative effect to better understand what the effect is.


From: SPSSX(r) Discussion on behalf of Omar Farook Sent: Wed 4/19/2006 12:03 AM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Re: effects of road diversions on shops

Dear Mr. Art, The sales and the number of customers could not increased because all the shops in the area are affected due to, non-availability of cars parking and some of internal roads are semi-closed. Now should I ignore the data or there is a way to deal with the four variables? Thanks. Omar.

Art Kendall <> wrote: It is biased because, if I understood your response format, the questioning did not allow for the possibility that sales or number of customers could have increased for some shops.

I don't know the wording and the language the study was done in. In English "affected" does not mean negatively affected. Even if the researcher could not get the number of customers, in comparing two points in time respondents could have been asked a questions that spanned more of the possible range of outcomes, something like :many more, some more, a few more, about the same, a few less, some less, many less.

The shop owners seemingly were asked the percentage decrease, not the relative percentage one way or the other.

Art Social Research Consultants

Omar Farook wrote:

>Dear ViAnn, > May I know why it is baize? > Also I would like to add that in our country it is quite difficult to ask about the amount of sales, you will not get anything at all. Also with respect to the number of customer, no one of them count the number of customers. > Thanks. > >"Beadle, ViAnn" wrote: If you are sure the questions asked were exactly what you state below, I agree with Art--your instrument is biased and any results from it are questionable. It sounds to me like somebody has a strong political agenda against road diversions and are trying to buttress that with "analysis". > >-----Original Message----- >From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Omar Farook >Sent: Tuesday, April 18, 2006 1:36 PM >To: SPSSX-L@LISTSERV.UGA.EDU >Subject: Re: effects of road diversions on shops > >Dear Mr. Art, >First of all I am not the person who created the questioner; my task is to analyze the data only. >We have 102 shops around the road works, 98 from them said the works affected my business and when they said yes we asked them the four questions which have been mentioned in my last message. >Many thanks. >Omar. > > >Art Kendall wrote: I assumed that he had a one tail hypothesis. That means that the z >-test that the average change result is different from zero would allow >a more lenient alpha. > >However, from the way I read his post, the way I assumed he asked the >question was "leading" in rhetorical terminology. Its demand >characteristics did not allow for results to go in either direction. >"how much more have you been beating your wife, compared to last month?" >if you only gather data where the values are on one side of zero (or of >100% for a percentage), then the summary of those values will be on one >side of zero. > >Based on the design criterion that the hypothesis be falsifiable, he >should have asked for the sales last month, and the sales now. If his >pretests found that people did not want to give him raw numbers, he >could back off to asking the respondents to use a separate worksheet he >provided and enter result on the survey instrument. > >As an aside he still would have the problem that percentages cannot >simply be averaged. The result summarizing the sample should calculate >the percent by summing the numerators for each case, and dividing by the >sum of the denominators. > > >Art > >Social Research Consultants > > > >Mark A Davenport MADAVENP wrote: > > > >>Art, >> >>Might he be talking about a uni-directional (one-tailed) hypothesis? I >>agree, the description is a bit hard to make heads or tails of (no pun >>intended). >> >>Mark >> >>*************************************************************************************************************************************************************** >>Mark A. Davenport Ph.D. >>Asst. to the Vice Chancellor for Student Affairs >>Office of Student Affairs Research and Evaluation >>The University of North Carolina at Greensboro >>336.334.5582 >> >> >>'An approximate answer to the right question is worth a good deal more >>than an exact answer to an approximate question.' --a paraphrase of J. W. >>Tukey (1962) >> >> >> >> >> >> >>Art Kendall >>Sent by: "SPSSX(r) Discussion" >>04/18/2006 08:58 AM >>Please respond to >> >> >> >>To >>SPSSX-L@LISTSERV.UGA.EDU >>cc >> >>Subject >>Re: effects of road diversions on shops >> >> >> >> >> >> >>If (a big if) I understand you correctly you are up a creek. You >>violated the design principle of having your hypothesis be falsifiable. >> >>You did not allow for the possibility that sales or customers could >>remain the same or even increase. >> >>If you had done that, you could have done boxplots. >> >>Art >> >>Social Research Consultants >> >> >>Omar Farook wrote: >> >> >> >> >> >>>Dear Experts. >>>I have 4 variables, in var1 the shop's owner should mention the >>> >>> >>> >>> >>decrease percentage in sales comparison with the month preceding the >>roads work .in var2 the shop's owner should mention the decrease >>percentage in sales also but comparison with the same month from the >>previous year. >> >> >> >> >>>in var3 the shop's owner should mention the decrease percentage in >>> >>> >>> >>> >>number of customers comparison with the month preceding the roads work .in >>var4 the shop's owner should mention the decrease percentage in number of >>customers comparison with the same month from the previous year. >> >> >> >> >>>May I know the statistical methods that might be used to analyze these >>> >>> >>> >>> >>variables? >> >> >> >> >>>Hope I was clear enough. >>>Thanks. >>>Omar. >>> >>> >>>--------------------------------- >>>Yahoo! Messenger with Voice. PC-to-Phone calls for ridiculously low >>> >>> >>> >>> >>rates. >> >> >> >> >>> >>> >>> >>> >> >> >> >> > > > >--------------------------------- >New Yahoo! Messenger with Voice. Call regular phones from your PC and save big. > > > >--------------------------------- >Love cheap thrills? Enjoy PC-to-Phone calls to 30+ countries for just 2/min with Yahoo! Messenger with Voice. > > > >

--------------------------------- Love cheap thrills? Enjoy PC-to-Phone calls to 30+ countries for just 2/min with Yahoo! Messenger with Voice.

Back to: Top of message | Previous page | Main SPSSX-L page