|
CraigM
|
|
|
Group: Forum Members
Posts: 14,
Visits: 70
|
Hi Dave and Sean.
I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7.
We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered.
We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders:
Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3.
Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3.
I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files.
This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes.
I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better.
Best wishes.
~Craig
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 109K
|
+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose?
|
|
|
|
|
CraigM
|
|
|
Group: Forum Members
Posts: 14,
Visits: 70
|
+x+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose? Thanks Dave. Maybe I can assign by group number. My first instinct was to use subject number since that’s how I’ve always done single-experiment scripts. I didn’t realize it didn’t work for batch scripts. We are testing elementary school students in their schools, on their school laptops, and without a researcher to facilitate. We are going to have all students use the same Inquisit Web study link, and then their teachers will enter the subject IDs into Inquisit for every student. Every student will complete three tests spread across three days, and the order of those tests will be counterbalanced. Within those constraints, I have not been able to find a way to use group number to help us. I’m not saying it’s impossible, but I have tried, and I can’t figure it out. One thing I can’t resolve is that, as far as I can tell, group number cannot be entered by the user (a teacher, in our case) on Inquisit Web. And I can’t include group number as a URL parameter because every student will be using the same link. I also can’t have group number be random or sequential because the group number needs to be the same for all three sessions within each student (i.e., if a student is assigned group 1 on Monday, it needs to be guaranteed that they’ll be assigned to group 1 on Wednesday and Friday). I’ve also tried using “/ groupAssignment = subjectNumber” to circumvent (a) the subjects attribute not working in a batch file and (b) not being able to manually user-enter group number on Inquisit Web. That caused Inquisit to run every test in the batch file one after another, instead of just stopping after the first test. But maybe I goofed somewhere there. My hope at this point is that I’ve overcomplicated things or just missed something obvious. Whether it’s that or this really is just a complicated problem to resolve in Inquisit, can you offer any guidance on how I can achieve what we’re going for based on the real-world testing scenario I described above? The only fallback idea I have is to combine my three tests into one script and use a bunch of “/ skip = [script.sessionId == #]” commands throughout, but I think that would be very time consuming (but I will do it if I have to). Thank you, I appreciate your time, Dave.
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 109K
|
+x+x+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose? Thanks Dave. Maybe I can assign by group number. My first instinct was to use subject number since that’s how I’ve always done single-experiment scripts. I didn’t realize it didn’t work for batch scripts. We are testing elementary school students in their schools, on their school laptops, and without a researcher to facilitate. We are going to have all students use the same Inquisit Web study link, and then their teachers will enter the subject IDs into Inquisit for every student. Every student will complete three tests spread across three days, and the order of those tests will be counterbalanced. Within those constraints, I have not been able to find a way to use group number to help us. I’m not saying it’s impossible, but I have tried, and I can’t figure it out. One thing I can’t resolve is that, as far as I can tell, group number cannot be entered by the user (a teacher, in our case) on Inquisit Web. And I can’t include group number as a URL parameter because every student will be using the same link. I also can’t have group number be random or sequential because the group number needs to be the same for all three sessions within each student (i.e., if a student is assigned group 1 on Monday, it needs to be guaranteed that they’ll be assigned to group 1 on Wednesday and Friday). I’ve also tried using “/ groupAssignment = subjectNumber” to circumvent (a) the subjects attribute not working in a batch file and (b) not being able to manually user-enter group number on Inquisit Web. That caused Inquisit to run every test in the batch file one after another, instead of just stopping after the first test. But maybe I goofed somewhere there. My hope at this point is that I’ve overcomplicated things or just missed something obvious. Whether it’s that or this really is just a complicated problem to resolve in Inquisit, can you offer any guidance on how I can achieve what we’re going for based on the real-world testing scenario I described above? The only fallback idea I have is to combine my three tests into one script and use a bunch of “/ skip = [script.sessionId == #]” commands throughout, but I think that would be very time consuming (but I will do it if I have to). Thank you, I appreciate your time, Dave. What I don't understand, and perhaps you can clarify: Having a teacher enter subject ID for every student seems no less complicated and error prone than having said teacher click / open a specific link for each student, which would already contain the subject ID assigned to said student, their assigned group number, and -- optionally -- their session ID as well. What am I missing?
|
|
|
|
|
CraigM
|
|
|
Group: Forum Members
Posts: 14,
Visits: 70
|
+x+x+x+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose? Thanks Dave. Maybe I can assign by group number. My first instinct was to use subject number since that’s how I’ve always done single-experiment scripts. I didn’t realize it didn’t work for batch scripts. We are testing elementary school students in their schools, on their school laptops, and without a researcher to facilitate. We are going to have all students use the same Inquisit Web study link, and then their teachers will enter the subject IDs into Inquisit for every student. Every student will complete three tests spread across three days, and the order of those tests will be counterbalanced. Within those constraints, I have not been able to find a way to use group number to help us. I’m not saying it’s impossible, but I have tried, and I can’t figure it out. One thing I can’t resolve is that, as far as I can tell, group number cannot be entered by the user (a teacher, in our case) on Inquisit Web. And I can’t include group number as a URL parameter because every student will be using the same link. I also can’t have group number be random or sequential because the group number needs to be the same for all three sessions within each student (i.e., if a student is assigned group 1 on Monday, it needs to be guaranteed that they’ll be assigned to group 1 on Wednesday and Friday). I’ve also tried using “/ groupAssignment = subjectNumber” to circumvent (a) the subjects attribute not working in a batch file and (b) not being able to manually user-enter group number on Inquisit Web. That caused Inquisit to run every test in the batch file one after another, instead of just stopping after the first test. But maybe I goofed somewhere there. My hope at this point is that I’ve overcomplicated things or just missed something obvious. Whether it’s that or this really is just a complicated problem to resolve in Inquisit, can you offer any guidance on how I can achieve what we’re going for based on the real-world testing scenario I described above? The only fallback idea I have is to combine my three tests into one script and use a bunch of “/ skip = [script.sessionId == #]” commands throughout, but I think that would be very time consuming (but I will do it if I have to). Thank you, I appreciate your time, Dave. What I don't understand, and perhaps you can clarify: Having a teacher enter subject ID for every student seems no less complicated and error prone than having said teacher click / open a specific link for each student, which would already contain the subject ID assigned to said student, their assigned group number, and -- optionally -- their session ID as well. What am I missing? District staff will be installing the Inquisit Player on every device and putting a shortcut to our study link on the computer desktop. This will make it so the teacher hardly has to do anything at the time of testing: enter an ID number from their clipboard and click start. Because we are likely going to be testing over 1,000 students, it is not reasonable for us to give the district a list of 1,000 links and expect them to find every student's unique link and put it on the correct device. Unfortunately, we are not permitted to get on the devices ourselves and take on that responsibility. We need everything to be as simple as possible, first, for the teachers because most of them either won't want to be dealing with the study at all or won't be able to follow any directions longer than a couple steps. After that, we want things to be as simple as possible for the district, too, because if it becomes too much work, they could just decide to say, "no, we can't do this study anymore." Lastly, we need to put as little responsibility on the students as possible because we cannot trust their ability to launch the study correctly or enter any ID numbers (some will be as young as 5 years old). We are extremely lucky to have this opportunity, so I am trying to take on all the responsibility and do all the heavy lifting so the district and schools don't have to do much work. It is certainly possible there will be some errors with manually entering subject numbers, but we are doing everything in our power to lower the chance as far as we can. I thought batch scripts were going to be the way to go, but if not, then I'll go with my labor intensive fallback option.
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 109K
|
+x+x+x+x+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose? Thanks Dave. Maybe I can assign by group number. My first instinct was to use subject number since that’s how I’ve always done single-experiment scripts. I didn’t realize it didn’t work for batch scripts. We are testing elementary school students in their schools, on their school laptops, and without a researcher to facilitate. We are going to have all students use the same Inquisit Web study link, and then their teachers will enter the subject IDs into Inquisit for every student. Every student will complete three tests spread across three days, and the order of those tests will be counterbalanced. Within those constraints, I have not been able to find a way to use group number to help us. I’m not saying it’s impossible, but I have tried, and I can’t figure it out. One thing I can’t resolve is that, as far as I can tell, group number cannot be entered by the user (a teacher, in our case) on Inquisit Web. And I can’t include group number as a URL parameter because every student will be using the same link. I also can’t have group number be random or sequential because the group number needs to be the same for all three sessions within each student (i.e., if a student is assigned group 1 on Monday, it needs to be guaranteed that they’ll be assigned to group 1 on Wednesday and Friday). I’ve also tried using “/ groupAssignment = subjectNumber” to circumvent (a) the subjects attribute not working in a batch file and (b) not being able to manually user-enter group number on Inquisit Web. That caused Inquisit to run every test in the batch file one after another, instead of just stopping after the first test. But maybe I goofed somewhere there. My hope at this point is that I’ve overcomplicated things or just missed something obvious. Whether it’s that or this really is just a complicated problem to resolve in Inquisit, can you offer any guidance on how I can achieve what we’re going for based on the real-world testing scenario I described above? The only fallback idea I have is to combine my three tests into one script and use a bunch of “/ skip = [script.sessionId == #]” commands throughout, but I think that would be very time consuming (but I will do it if I have to). Thank you, I appreciate your time, Dave. What I don't understand, and perhaps you can clarify: Having a teacher enter subject ID for every student seems no less complicated and error prone than having said teacher click / open a specific link for each student, which would already contain the subject ID assigned to said student, their assigned group number, and -- optionally -- their session ID as well. What am I missing? District staff will be installing the Inquisit Player on every device and putting a shortcut to our study link on the computer desktop. This will make it so the teacher hardly has to do anything at the time of testing: enter an ID number from their clipboard and click start. Because we are likely going to be testing over 1,000 students, it is not reasonable for us to give the district a list of 1,000 links and expect them to find every student's unique link and put it on the correct device. Unfortunately, we are not permitted to get on the devices ourselves and take on that responsibility. We need everything to be as simple as possible, first, for the teachers because most of them either won't want to be dealing with the study at all or won't be able to follow any directions longer than a couple steps. After that, we want things to be as simple as possible for the district, too, because if it becomes too much work, they could just decide to say, "no, we can't do this study anymore." Lastly, we need to put as little responsibility on the students as possible because we cannot trust their ability to launch the study correctly or enter any ID numbers (some will be as young as 5 years old). We are extremely lucky to have this opportunity, so I am trying to take on all the responsibility and do all the heavy lifting so the district and schools don't have to do much work. It is certainly possible there will be some errors with manually entering subject numbers, but we are doing everything in our power to lower the chance as far as we can. I thought batch scripts were going to be the way to go, but if not, then I'll go with my labor intensive fallback option. Understood. Let me think a little about how to make this work just with subject ID for counterbalancing.
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 109K
|
+x+x+x+x+x+xHi Dave and Sean. I am having trouble with a batch file that uses the subjects and sessions attributes. I am uploading some files that I am using just to test out the process. I am using Inquisit 7. We are going to be testing participants remotely using Inquisit Web. Each participant will complete three tests total, with each test being completed on a different session on a different day. Due to constraints of our testing scenario, we will be using one general link for all participants and subject ID will be user entered. We want to counterbalance the order of the three tests across participants. For the sake of this demo I made for myself, I only have two test orders: Odd numbered subject IDs would complete Test A at Session 1, Test B at Session 2, and Test C at Session 3. Even numbered subject IDs would complete Test C at Session 1, Test A at Session 2, and Test B at Session 3. I thought I had programmed this correctly in the attached files. I have one batch file that should select the appropriate test based on the subject ID and session ID, and then three individual test files. This is working as intended for subject IDs 1 and 2 at session IDs 1, 2, and 3. But it does not work for subject IDs greater than 2 or session IDs greater than 3; in those cases, Inquisit just says "The session is complete" without running a test. I can't figure out why this batch file is not acknowledging the modulus number in the subject ID and session ID attributes. I have read a few similar topics on the forums, and I have also tried a batch nesting method, but I am at a loss. Can you help me find my error or offer a different strategy? Please let me know if you need me to explain anything better. Best wishes. ~Craig Assignment by subject ID is strongly discouraged and not really supported anymore. Reason being that subject IDs in the web context typically aren't numerical, which researhers tend to overlook, and then run into problems when they deploy their tasks online. In other words: In practice assignment by subject ID causes more issues than it solves. Why can't you assign by group number, which was introduced exactly for this purpose? Thanks Dave. Maybe I can assign by group number. My first instinct was to use subject number since that’s how I’ve always done single-experiment scripts. I didn’t realize it didn’t work for batch scripts. We are testing elementary school students in their schools, on their school laptops, and without a researcher to facilitate. We are going to have all students use the same Inquisit Web study link, and then their teachers will enter the subject IDs into Inquisit for every student. Every student will complete three tests spread across three days, and the order of those tests will be counterbalanced. Within those constraints, I have not been able to find a way to use group number to help us. I’m not saying it’s impossible, but I have tried, and I can’t figure it out. One thing I can’t resolve is that, as far as I can tell, group number cannot be entered by the user (a teacher, in our case) on Inquisit Web. And I can’t include group number as a URL parameter because every student will be using the same link. I also can’t have group number be random or sequential because the group number needs to be the same for all three sessions within each student (i.e., if a student is assigned group 1 on Monday, it needs to be guaranteed that they’ll be assigned to group 1 on Wednesday and Friday). I’ve also tried using “/ groupAssignment = subjectNumber” to circumvent (a) the subjects attribute not working in a batch file and (b) not being able to manually user-enter group number on Inquisit Web. That caused Inquisit to run every test in the batch file one after another, instead of just stopping after the first test. But maybe I goofed somewhere there. My hope at this point is that I’ve overcomplicated things or just missed something obvious. Whether it’s that or this really is just a complicated problem to resolve in Inquisit, can you offer any guidance on how I can achieve what we’re going for based on the real-world testing scenario I described above? The only fallback idea I have is to combine my three tests into one script and use a bunch of “/ skip = [script.sessionId == #]” commands throughout, but I think that would be very time consuming (but I will do it if I have to). Thank you, I appreciate your time, Dave. What I don't understand, and perhaps you can clarify: Having a teacher enter subject ID for every student seems no less complicated and error prone than having said teacher click / open a specific link for each student, which would already contain the subject ID assigned to said student, their assigned group number, and -- optionally -- their session ID as well. What am I missing? District staff will be installing the Inquisit Player on every device and putting a shortcut to our study link on the computer desktop. This will make it so the teacher hardly has to do anything at the time of testing: enter an ID number from their clipboard and click start. Because we are likely going to be testing over 1,000 students, it is not reasonable for us to give the district a list of 1,000 links and expect them to find every student's unique link and put it on the correct device. Unfortunately, we are not permitted to get on the devices ourselves and take on that responsibility. We need everything to be as simple as possible, first, for the teachers because most of them either won't want to be dealing with the study at all or won't be able to follow any directions longer than a couple steps. After that, we want things to be as simple as possible for the district, too, because if it becomes too much work, they could just decide to say, "no, we can't do this study anymore." Lastly, we need to put as little responsibility on the students as possible because we cannot trust their ability to launch the study correctly or enter any ID numbers (some will be as young as 5 years old). We are extremely lucky to have this opportunity, so I am trying to take on all the responsibility and do all the heavy lifting so the district and schools don't have to do much work. It is certainly possible there will be some errors with manually entering subject numbers, but we are doing everything in our power to lower the chance as far as we can. I thought batch scripts were going to be the way to go, but if not, then I'll go with my labor intensive fallback option. Understood. Let me think a little about how to make this work just with subject ID for counterbalancing. Alright, the best option I can think of is to use condtional <include> elements for the counterbalancing, i.e. dynamically load in the applicable task script dependng on whether the provided subject ID is odd or even. The batch script, then, would only assign by session number to a stub script for the given day. Working example based on your test scripts is attached below.
|
|
|
|
|
CraigM
|
|
|
Group: Forum Members
Posts: 14,
Visits: 70
|
This is wonderful, thanks Dave. I tried it out and it does exactly what we need it to. I'm confident I can scale it up and apply it to our actual tests.
It still runs into the issue of not working for session numbers greater than 3, but that is easy to get around. In a perfect world, we would not need to worry about session numbers >3, but I'm trying to build a safety net where it does work just for some unlikely scenarios (not worth going into). But all I have to do is change the sessions attribute in the start file so "(1, 4, 7 of 9)" all lead to the session 1 task, and so on. That will be easy, and that will give us a buffer in case any of those unlikely scenarios occur.
Thanks Dave. I truly appreciate your time and effort in putting this solution together for us.
~Craig
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 109K
|
+xThis is wonderful, thanks Dave. I tried it out and it does exactly what we need it to. I'm confident I can scale it up and apply it to our actual tests. It still runs into the issue of not working for session numbers greater than 3, but that is easy to get around. In a perfect world, we would not need to worry about session numbers >3, but I'm trying to build a safety net where it does work just for some unlikely scenarios (not worth going into). But all I have to do is change the sessions attribute in the start file so "(1, 4, 7 of 9)" all lead to the session 1 task, and so on. That will be easy, and that will give us a buffer in case any of those unlikely scenarios occur. Thanks Dave. I truly appreciate your time and effort in putting this solution together for us. ~Craig What I would suggest is to ask the admins to put three links on the machines, for Day 1, Day 2, and Day 3 respectively. I.e. suppose the URL for your study were https://mili2nd.co/abcd , then: Day 1Day 2Day 3That way you can always be sure that the correct session / task is launched, even if something goes wrong, and somebody has to restart / try again.
|
|
|
|
|
CraigM
|
|
|
Group: Forum Members
Posts: 14,
Visits: 70
|
+x+xThis is wonderful, thanks Dave. I tried it out and it does exactly what we need it to. I'm confident I can scale it up and apply it to our actual tests. It still runs into the issue of not working for session numbers greater than 3, but that is easy to get around. In a perfect world, we would not need to worry about session numbers >3, but I'm trying to build a safety net where it does work just for some unlikely scenarios (not worth going into). But all I have to do is change the sessions attribute in the start file so "(1, 4, 7 of 9)" all lead to the session 1 task, and so on. That will be easy, and that will give us a buffer in case any of those unlikely scenarios occur. Thanks Dave. I truly appreciate your time and effort in putting this solution together for us. ~Craig What I would suggest is to ask the admins to put three links on the machines, for Day 1, Day 2, and Day 3 respectively. I.e. suppose the URL for your study were https://mili2nd.co/abcd , then: Day 1Day 2Day 3That way you can always be sure that the correct session / task is launched, even if something goes wrong, and somebody has to restart / try again. That is a good suggestion, thank you. That will be my first choice if the district admins will go for it.
|
|
|
|