leezustheboy
Badges: 2
Rep:
?
#1
Report Thread starter 6 months ago
#1
Hi,

Just wondering as to what statistical tests I should use to see a correlation in these aims for my diss.

My aims are:

Association of salt thresholds (different concentrations) and dietary intake (kcal)

Association between preference for salty food (Sparingly, Average and Salty categories) and dietary intake (kcal).


I need this ASAP so any help would be massively appreciated as I am shambolic at statistics.

Thanks in advance!!!
0
reply
mqb2766
Badges: 17
Rep:
?
#2
Report 6 months ago
#2
Linear correlation or rank or ...
What does the data look like?
(Original post by leezustheboy)
Hi,

Just wondering as to what statistical tests I should use to see a correlation in these aims for my diss.

My aims are:

Association of salt thresholds (different concentrations) and dietary intake (kcal)

Association between preference for salty food (Sparingly, Average and Salty categories) and dietary intake (kcal).


I need this ASAP so any help would be massively appreciated as I am shambolic at statistics.

Thanks in advance!!!
0
reply
leezustheboy
Badges: 2
Rep:
?
#3
Report Thread starter 6 months ago
#3
Thanks for the reply!

The data for the thresholds is; recognition threshold and detection threshold
then i want to see if there is relationship between high thresholds and dietary intake (which is just kcal from 24-hr recall) which i think is linear (I'm not sure).

For the other aim (salty preference) the 3 categories were converted to numbers (1=sparingly, 2=average, 3=salty) then dietary intake is same as above (24hr-recall)


this the threshold data

Name:  Screen Shot 2019-04-05 at 09.44.59.png
Views: 12
Size:  33.1 KB
this is the salt preference and dietary intake

Name:  Screen Shot 2019-04-05 at 09.46.39.png
Views: 14
Size:  9.2 KB
0
reply
mqb2766
Badges: 17
Rep:
?
#4
Report 6 months ago
#4
Do a correlation plot to eyeball the data?
(Original post by leezustheboy)
Thanks for the reply!

The data for the thresholds is; recognition threshold and detection threshold
then i want to see if there is relationship between high thresholds and dietary intake (which is just kcal from 24-hr recall) which i think is linear (I'm not sure).

For the other aim (salty preference) the 3 categories were converted to numbers (1=sparingly, 2=average, 3=salty) then dietary intake is same as above (24hr-recall)


this the threshold data

Name:  Screen Shot 2019-04-05 at 09.44.59.png
Views: 12
Size:  33.1 KB
this is the salt preference and dietary intake

Name:  Screen Shot 2019-04-05 at 09.46.39.png
Views: 14
Size:  9.2 KB
0
reply
leezustheboy
Badges: 2
Rep:
?
#5
Report Thread starter 6 months ago
#5
(Original post by mqb2766)
Do a correlation plot to eyeball the data?
I tried to put the threshold averages and dietary intake into the correlation plot with dietary intake but it had an error in excel...this the the threshold and recognition together if that helps?

Name:  Screen Shot 2019-04-05 at 10.49.28.png
Views: 12
Size:  6.1 KB
0
reply
mqb2766
Badges: 17
Rep:
?
#6
Report 6 months ago
#6
You really need to do a scatter plot of the two variables you're interested in. Try and get rid of the error?
(Original post by leezustheboy)
I tried to put the threshold averages and dietary intake into the correlation plot with dietary intake but it had an error in excel...this the the threshold and recognition together if that helps?

Name:  Screen Shot 2019-04-05 at 10.49.28.png
Views: 12
Size:  6.1 KB
0
reply
leezustheboy
Badges: 2
Rep:
?
#7
Report Thread starter 6 months ago
#7
Would it be worth using two variables instead of three? As I have been trying to put detection, recognition and dietary intake together? Instead I could use recognition (when someone can identify the taste as salt) and dietary intake to see the association between high recognition thresholds and dietary intake?
(Original post by mqb2766)
You really need to do a scatter plot of the two variables you're interested in. Try and get rid of the error?
0
reply
mqb2766
Badges: 17
Rep:
?
#8
Report 6 months ago
#8
Post 1 seemed to suggest you were looking at two correlation problems, each involving two variables?
(Original post by leezustheboy)
Would it be worth using two variables instead of three? As I have been trying to put detection, recognition and dietary intake together? Instead I could use recognition (when someone can identify the taste as salt) and dietary intake to see the association between high recognition thresholds and dietary intake?
0
reply
leezustheboy
Badges: 2
Rep:
?
#9
Report Thread starter 6 months ago
#9
Yes there are two variables but the high thresholds is split into two groups being detection and recognition with the second variable being dietary intake. Similarly, the preference for salt is split into 3 groups/catergories being salty, sparingly and average with the second variable being dietary intake. I'm trying to show a relationship with higher salt recognition leads to higher dietary intake so those with that higher intake should have higher detection and recognition thresholds. Additionally, for the other test those that have a preference for the saltier foods should have a higher dietary intake. The splitting of the variables is confusing me as I'm trying to show a relationship but seems like I have more than 2 variables?
(Original post by mqb2766)
Post 1 seemed to suggest you were looking at two correlation problems, each involving two variables?
0
reply
mqb2766
Badges: 17
Rep:
?
#10
Report 6 months ago
#10
I can't really help with the meaning of the variables.

(Original post by leezustheboy)
Yes there are two variables but the high thresholds is split into two groups being detection and recognition with the second variable being dietary intake. Similarly, the preference for salt is split into 3 groups/catergories being salty, sparingly and average with the second variable being dietary intake. I'm trying to show a relationship with higher salt recognition leads to higher dietary intake so those with that higher intake should have higher detection and recognition thresholds. Additionally, for the other test those that have a preference for the saltier foods should have a higher dietary intake. The splitting of the variables is confusing me as I'm trying to show a relationship but seems like I have more than 2 variables?
0
reply
leezustheboy
Badges: 2
Rep:
?
#11
Report Thread starter 6 months ago
#11
I think it is a linear regression for the detection and recognition thresholds then instead of using dietary intake use preference for salty taste?


Then a separate test to test whether different preferences for salty taste relate to dietary intake?
(Original post by mqb2766)
I can't really help with the meaning of the variables.
0
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

University open days

  • University of Surrey
    Postgraduate Open Afternoon Postgraduate
    Wed, 23 Oct '19
  • University of Bristol
    Undergraduate Open Afternoon Undergraduate
    Wed, 23 Oct '19
  • University of Exeter
    Undergraduate Open Day - Penryn Campus Undergraduate
    Wed, 23 Oct '19

Would you turn to a teacher if you were being bullied?

Yes (11)
21.15%
No (41)
78.85%

Watched Threads

View All