spagetti Plot for a Discrete Choice Experiement

[Lynette LeKrampf]

Hello, I am very new to Stata and am researching if someone would self-identify as being ill from a publicized foodborne illness or not and why. I collected data using a Discrete Choice Experiment. I analyzed my data using a mixed logit model. I have four attributes: the number of people (3 levels at 20, 200, 8500), symptoms (2 levels, included or not), FDA (2 levels, included or not), and a call to Action (2 levels, included or not). Each person was given six choice set questions. My data is set up the long way (example below with 2 participants data). I used effects coding on the 20 people level, omitting the 20). How do I use STATA to create a Spaghetti Plot?

Participant_ID	Group	Choice_Set_ID	Alternative	Chosen	People	Symptoms	FDA
1	1	Block_a_choice Set_1_Response	A	1	85	1	1
1	1	Block_a_choice Set_1_Response	B	0	0.2	-1	-1
1	1	Block_a_choice Set_1_Response	Neither	0	0	0	0
1	5	Block_a_choice Set_5_Response	A	1	85	-1	-1
1	5	Block_a_choice Set_5_Response	B	0	0.2	1	1
1	5	Block_a_choice Set_5_Response	Neither	0	0	0	0
1	9	Block_a_choice Set_9_Response	A	1	2	1	1
1	9	Block_a_choice Set_9_Response	B	0	85	-1	-1
1	9	Block_a_choice Set_9_Response	Neither	0	0	0	0
1	13	Block_a_choice Set_13_Response	A	0	2	-1	-1
1	13	Block_a_choice Set_13_Response	B	1	85	1	1
1	13	Block_a_choice Set_13_Response	Neither	0	0	0	0
1	17	Block_a_choice Set_17_Response	A	1	2	1	1
1	17	Block_a_choice Set_17_Response	B	0	0.2	-1	-1
1	17	Block_a_choice Set_17_Response	Neither	0	0	0	0
1	21	Block_a_choice Set_21_Response	A	1	2	-1	-1
1	21	Block_a_choice Set_21_Response	B	0	0.2	1	1
1	21	Block_a_choice Set_21_Response	Neither	0	0	0	0
2	2	Block_b_choiceSet_2_response	A	1	85	1	1
2	2	Block_b_choiceSet_2_response	B	0	0.2	-1	-1
2	2	Block_b_choiceSet_2_response	Neither	0	0	0	0
2	6	Block_b_choiceSet_6_response	A	0	85	-1	-1
2	6	Block_b_choiceSet_6_response	B	1	0.2	1	1
2	6	Block_b_choiceSet_6_response	Neither	0	0	0	0
2	10	Block_b_choiceSet_10_response	A	1	2	1	1
2	10	Block_b_choiceSet_10_response	B	0	85	-1	-1
2	10	Block_b_choiceSet_10_response	Neither	0	0	0	0
2	14	Block_b_choiceSet_14_response	A	0	2	-1	-1
2	14	Block_b_choiceSet_14_response	B	1	85	1	1
2	14	Block_b_choiceSet_14_response	Neither	0	0	0	0
2	18	Block_b_choiceSet_18_response	A	1	2	1	1
2	18	Block_b_choiceSet_18_response	B	0	0.2	-1	-1
2	18	Block_b_choiceSet_18_response	Neither	0	0	0	0
2	22	Block_b_choiceSet_22_response	A	0	2	-1	-1
2	22	Block_b_choiceSet_22_response	B	1	0.2	1	1
2	22	Block_b_choiceSet_22_response	Neither	0	0	0	0

do

[Lynette LeKrampf]

Spaghetti Plot... (way to make an impression on my first post!) Thank you for the GRACE!

[Nick Cox]

https://www.statalist.org/forums/help#spelling

More seriously, what goes on the y axis? what on the x axis? What is repeated within each graphical panel (facet)? What is separated in different panels (facets)?

[Lynette LeKrampf]

After more research, I found an example of what I hope to create in other research (A discrete choice experiment for attribute preferences of androgen deprivation therapies in prostate cancer by Hauber et al., 2024). Now I have to figure out how to create this with my results.

[Nick Cox]

Sorry, but I am still on the dark on what your answers are to #3. In addition it is hard to make sense of 00.2 185 and the like as supposedly values of the same variable. Please see https://www.statalist.org/forums/help#stata on how to present a data example.

Bottom lines include:

Line plots require twoway line or twoway connect.

Either of those in turn requires that what is plotted on the y and x axes are numeric variables.

The paper you mention can be found at https://link.springer.com/article/10...25-024-02955-1 I can't agree myself that the display you show is any kind of exemplar to be emulated.