One likert scale question, 3 hypotheses analysis help needed
Locked
One likert scale question, 3 hypotheses analysis help needed
 
|
I think a VERY simple analysis question but I am at a lost as a novice to
statistics AND SPSS 🙁 We did a test on three homogenic groups (test A, B and C) (people in the group are homogenic as well as the groups compared to each other) of 100 students each to find out if packaging can influence the taste of a certain kind of candy. We know all the students do not normally like this type of candy. Each individual was asked to look closely at the packaging and after tasting the candy to point to a 5 point likert scale from 1 to 5 ((1)Very Good, Good, OK, Bad, (5) Very Bad). We have 3 hypotheses: 1. We think packaging used in test A will have no effect on the taste experience 2. We think packaging used in test B will have little positive effect on the taste experience 2. We think packaging used in test C will have significant positive effect on the taste experience. EXAMPLE: the data looks like this. Meaning test A we 5 likert scale results of which 1 was Very Good and 3 was Good and 1 was OK, etc). In reality we have 370 results. A = 1 A = 2 A = 2 A = 2 A = 3 B = 2 B = 2 B = 3 B = 3 The data is already in SPSS the A, B and C values are actually numbers "ofcourse" in de data view tab. I have two variables set. TEST Numeric 8 0 {1, CNTEST}... None 11 Right Ordinal Input TASTE Numeric 8 0 {1, Very Good}... None 11 Right Ordinal Input What is the best analyses to use on this. A few pointers would be very VERY appreciated. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: One likert scale question, 3 hypotheses analysis help needed
|
|
From the weaknesses of design, I figure that this must be a student's design for an assignment.
How do you test a likert-type variable? - usually, as if it were continuous, equal-interval, and normally distributed.
How do you test three groups on one continuous variable? - ANOVA.
How do you test an "ordered-groups" hypothesis? - Do a linear contrast across the groups, which is very similar to (if it is not completely the same test as) the correlation between the two variables; or do a t-test
between the extreme groups.
How do you improve the design? - There is much more power for simple comparisons (like this one) when one subject does multiple ratings.
There is very little burden on a subject tasting candies to do more than one rating, but showing the same candy in three wrappings is not a great idea. I would suggest, Let each student rate three different candies,
with randomized order of wrappings. By the way, "all the students do not normally like this type of candy" seems very unlikely, or the candy has a major marketing problem.
-- Rich Ulrich
From: SPSSX(r) Discussion <[hidden email]> on behalf of shardphoenix <[hidden email]>
Sent: Thursday, June 7, 2018 7:11:44 AM To: [hidden email] Subject: One likert scale question, 3 hypotheses analysis help needed I think a VERY simple analysis question but I am at a lost as a novice to
statistics AND SPSS 🙁 We did a test on three homogenic groups (test A, B and C) (people in the group are homogenic as well as the groups compared to each other) of 100 students each to find out if packaging can influence the taste of a certain kind of candy. We know all the students do not normally like this type of candy. Each individual was asked to look closely at the packaging and after tasting the candy to point to a 5 point likert scale from 1 to 5 ((1)Very Good, Good, OK, Bad, (5) Very Bad). We have 3 hypotheses: 1. We think packaging used in test A will have no effect on the taste experience 2. We think packaging used in test B will have little positive effect on the taste experience 2. We think packaging used in test C will have significant positive effect on the taste experience. EXAMPLE: the data looks like this. Meaning test A we 5 likert scale results of which 1 was Very Good and 3 was Good and 1 was OK, etc). In reality we have 370 results. A = 1 A = 2 A = 2 A = 2 A = 3 B = 2 B = 2 B = 3 B = 3 The data is already in SPSS the A, B and C values are actually numbers "ofcourse" in de data view tab. I have two variables set. TEST Numeric 8 0 {1, CNTEST}... None 11 Right Ordinal Input TASTE Numeric 8 0 {1, Very Good}... None 11 Right Ordinal Input What is the best analyses to use on this. A few pointers would be very VERY appreciated. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: One likert scale question, 3 hypotheses analysis help needed
|
We need to filter out student assignments, especially from MBA and similar students. John F Hall MA (Cantab) Dip Ed (Dunelm) [Retired academic survey researcher] Email: [hidden email] Website: Journeys in Survey Research Course: Survey Analysis Workshop (SPSS) From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Rich Ulrich From the weaknesses of design, I figure that this must be a student's design for an assignment. How do you test a likert-type variable? - usually, as if it were continuous, equal-interval, and normally distributed. How do you test three groups on one continuous variable? - ANOVA. How do you test an "ordered-groups" hypothesis? - Do a linear contrast across the groups, which is very similar to (if it is not completely the same test as) the correlation between the two variables; or do a t-test between the extreme groups. How do you improve the design? - There is much more power for simple comparisons (like this one) when one subject does multiple ratings. There is very little burden on a subject tasting candies to do more than one rating, but showing the same candy in three wrappings is not a great idea. I would suggest, Let each student rate three different candies, with randomized order of wrappings. By the way, "all the students do not normally like this type of candy" seems very unlikely, or the candy has a major marketing problem. -- Rich Ulrich From: SPSSX(r) Discussion <[hidden email]> on behalf of shardphoenix <[hidden email]> I think a VERY simple analysis question but I am at a lost as a novice to ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
I'm with you, John. I teach research methods, measurement, and statistics. The university has very strict rules governing ethics. Although the ethics policy is very long, the statement referring students to the policy ends with "Cheating, collusion, misconduct, fabrication, and plagiarism are considered serious offenses and violations can result in penalties up to and including expulsion from the University." On the other hand, I'm not sure that sometimes we don't have a student who's doing a directed study or internship and is free to ask resources, like the listserv, for assistance, that doesn't violate any ethical considerations.
Brian From: SPSSX(r) Discussion <[hidden email]> on behalf of John F Hall <[hidden email]>
Sent: Sunday, June 10, 2018 2:46:25 PM To: [hidden email] Subject: Re: One likert scale question, 3 hypotheses analysis help needed We need to filter out student assignments, especially from MBA and similar students.
John F Hall MA (Cantab) Dip Ed (Dunelm) [Retired academic survey researcher]
Email: [hidden email] Website: Journeys in Survey Research Course: Survey Analysis Workshop (SPSS)
From: SPSSX(r) Discussion <[hidden email]>
On Behalf Of Rich Ulrich
From the weaknesses of design, I figure that this must be a student's design for an assignment.
How do you test a likert-type variable? - usually, as if it were continuous, equal-interval, and normally distributed. How do you test three groups on one continuous variable? - ANOVA. How do you test an "ordered-groups" hypothesis? - Do a linear contrast across the groups, which is very similar to (if it is not completely the same test as) the correlation between the two variables; or do a t-test between the extreme groups.
How do you improve the design? - There is much more power for simple comparisons (like this one) when one subject does multiple ratings.
There is very little burden on a subject tasting candies to do more than one rating, but showing the same candy in three wrappings is not a great idea. I would suggest, Let each student rate three different candies, with randomized order of wrappings. By the way, "all the students do not normally like this type of candy" seems very unlikely, or the candy has a major marketing problem. -- Rich Ulrich
From: SPSSX(r) Discussion <[hidden email]> on behalf of shardphoenix <[hidden email]>
I think a VERY simple analysis question but I am at a lost as a novice to ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: One likert scale question, 3 hypotheses analysis help needed
Administrator
|
In reply to this post by John F Hall
Feel free to step up and moderate the group!
Enough SPAM to choke a horse. Feel free to ignore the homework assignments. Click/delete!!! John F Hall wrote > We need to filter out student assignments, especially from MBA and similar > students. > > > > John F Hall MA (Cantab) Dip Ed (Dunelm) > > [Retired academic survey researcher] > > > > Email: <mailto: > johnfhall@ > > > johnfhall@ > > > Website: <http://surveyresearch.weebly.com/> Journeys in Survey > Research > > Course: > <http://surveyresearch.weebly.com/1-survey-analysis-workshop-spss.html> > Survey Analysis Workshop (SPSS) > > Research: > <http://surveyresearch.weebly.com/3-subjective-social-indicators-quality-of-life.html> > Subjective Social Indicators (Quality of Life) > > > > From: SPSSX(r) Discussion < > SPSSX-L@.UGA > > On Behalf Of Rich Ulrich > Sent: 10 June 2018 19:00 > To: > SPSSX-L@.UGA > Subject: Re: One likert scale question, 3 hypotheses analysis help needed > > > > From the weaknesses of design, I figure that this must be a student's > design for an assignment. > > > > How do you test a likert-type variable? - usually, as if it were > continuous, equal-interval, and normally distributed. > > How do you test three groups on one continuous variable? - ANOVA. > > How do you test an "ordered-groups" hypothesis? - Do a linear contrast > across the groups, which is very similar to (if it is not completely the > same test as) the correlation between the two variables; or do a t-test > between the extreme groups. > > > > How do you improve the design? - There is much more power for simple > comparisons (like this one) when one subject does multiple ratings. > > > > There is very little burden on a subject tasting candies to do more than > one rating, but showing the same candy in three wrappings is not a great > idea. I would suggest, Let each student rate three different candies, with > randomized order of wrappings. By the way, "all the students do not > normally like this type of candy" seems very unlikely, or the candy has a > major marketing problem. > > -- > > Rich Ulrich > > > > _____ > > From: SPSSX(r) Discussion < > SPSSX-L@.UGA > <mailto: > SPSSX-L@.UGA > > > on behalf of shardphoenix < > shardphoenix@ > <mailto: > shardphoenix@ > > > > Sent: Thursday, June 7, 2018 7:11:44 AM > To: > SPSSX-L@.UGA > <mailto: > SPSSX-L@.UGA > > > Subject: One likert scale question, 3 hypotheses analysis help needed > > > > I think a VERY simple analysis question but I am at a lost as a novice to > statistics AND SPSS 🙁 > > We did a test on three homogenic groups (test A, B and C) (people in the > group are homogenic as well as the groups compared to each other) of 100 > students each to find out if packaging can influence the taste of a > certain > kind of candy. > We know all the students do not normally like this type of candy. > > Each individual was asked to look closely at the packaging and after > tasting > the candy to point to a 5 point likert scale from 1 to 5 ((1)Very Good, > Good, OK, Bad, (5) Very Bad). > > We have 3 hypotheses: > 1. We think packaging used in test A will have no effect on the taste > experience > 2. We think packaging used in test B will have little positive effect on > the > taste experience > 2. We think packaging used in test C will have significant positive effect > on the taste experience. > > EXAMPLE: the data looks like this. Meaning test A we 5 likert scale > results > of which 1 was Very Good and 3 was Good and 1 was OK, etc). In reality we > have 370 results. > > A = 1 > A = 2 > A = 2 > A = 2 > A = 3 > B = 2 > B = 2 > B = 3 > B = 3 > > The data is already in SPSS the A, B and C values are actually numbers > "ofcourse" in de data view tab. > I have two variables set. > > TEST Numeric 8 0 {1, CNTEST}... None 11 > Right Ordinal Input > > TASTE Numeric 8 0 {1, Very Good}... None 11 > Right Ordinal Input > > What is the best analyses to use on this. > A few pointers would be very VERY appreciated. > > > > -- > Sent from: http://spssx-discussion.1045642.n5.nabble.com/ > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > <mailto: > LISTSERV@.UGA > > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD > > ===================== To manage your subscription to SPSSX-L, send a > message to > LISTSERV@.UGA > <mailto: > LISTSERV@.UGA > > (not to SPSSX-L), with no body text except the command. To leave the > list, send the command SIGNOFF SPSSX-L For a list of commands to manage > subscriptions, send the command INFO REFCARD > > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- Please reply to the list and not to my personal email. Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
| Free forum by Nabble | Edit this page |