How to interpret residual covariances in lavaan

I generating a PA model using lavaan, so I would like to evaluate my model using fit indexes but residuals too.

Reading about I found this example:


# The Holzinger and Swineford (1939) example
HS.model <- ' visual  =~ x1 + x2 + x3
textual =~ x4 + x5 + x6
speed   =~ x7 + x8 + x9 '

fit <- lavaan(HS.model, data=HolzingerSwineford1939,
              auto.var=TRUE, auto.fix.first=TRUE,
summary(fit, fit.measures=TRUE)

# View Residuals (documentation in the lavaan-class help file)
resid(fit, type='normalized')

and the result was:
[1] "normalized"

   x1     x2     x3     x4     x5     x6     x7     x8     x9    
x1  0.000                                                        
x2 -0.493  0.000                                                 
x3 -0.125  1.539  0.000                                          
x4  1.159 -0.214 -1.170  0.000                                   
x5 -0.153 -0.459 -2.606  0.070  0.000                            
x6  0.983  0.507 -0.436 -0.130  0.048  0.000                     
x7 -2.423 -3.273 -1.450  0.625 -0.617 -0.240  0.000              
x8 -0.655 -0.896 -0.200 -1.162 -0.624 -0.375  1.170  0.000       
x9  2.405  1.249  2.420  0.808  1.126  0.958 -0.625 -0.504  0.000

x1 x2 x3 x4 x5 x6 x7 x8 x9 
 0  0  0  0  0  0  0  0  0 

How would I interpret these? Would somebody send me a reference? I am looking for an interpretation of this in lavaan and could not find good material.

How should I interpret information about bond spreads that narrowed or widened the most?

How should I interpret information about bond spreads that narrowed or widened the most? I see this information in the financial pages in the newspaper, and have some ideas about what it means, but have yet to find any definitive answers.

How to interpret mDNS query responses? [on hold]

I’m running Wireshark on a remote network for research purposes and seeing MDNS queries and responses. I understand the purpose of MDNS but have no idea how to interpret MDNS query responses.

Please find the screen shot below.

Is the domain name found at the bottom of the answer the host name of the local machine that sends this query response ? If so what is _services._dns-sd._udp.local ?
I don’t have the entire reply as the packet is truncated. Am I missing important information on the truncated part ?

Kindly let me know how to interpret this. If you could point me to a resource on interpreting this it would be very helpful.

enter image description here

How to interpret Mac’s dance in “It’s Always Sunny…”?

This article gives a glowing interpretation of Mac’s dance in the 13th season finale of It’s Always Sunny in Philadelphia. I don’t have access to the show, but I did watch the clip.

The article says it was about how he felt after coming out to his father. Perhaps it’s obvious to those who follow the show, but I just don’t get it.

My exposure to ballet and interpretive dance are limited, but his dancing with this woman doesn’t seem to express anything about his homosexuality. He was in close contact with her and performed impressive maneuvers, especially that downright heroic lift at 3:40. (Dances like this are what lead young straight boys into taking up ballet!). He didn’t display anything like distaste, discomfort, or even disinterest.

If he was trying to show his attraction to men, why didn’t he dance with a man?

If he was trying to show the pain of being rejected by his father (the man in the audience who got up and left, I presume), why was there no “character” representing the father?

If it was simply supposed to be an expression of his pain and vulnerability — well, most of those moves showed very little of that, except at the end.

What am I missing?

How to interpret grouping result in Tukey’s HSD test

Assuming there are three methods to kill bugs, and we want to know whether their performance differs. The Tukey’s HSD test gives the following grouping results:

  trt means  M
1   1  1.96  a
2   3  1.88 ab
3   2  1.68  b

If we need suggest some methods to use, can we say method 1 and 3 can both be used as they are either better than or equal to method 2 (both have the grouping letter ‘a’)?

How to interpret logarithmically transformed coefficients in linear regression?

My situation is:

I have 1 continuous dependent and 1 continuous predictor variable that I’ve logarithmically transformed to normalise their residuals for simple linear regression.

I would appreciate any help on how I can relate these transformed variables to their original context.

I want to use a linear regression to predict the number of days that pupils missed school in 2011 based on the number of days they missed in 2010. Most pupils miss 0 days or just a few days the data is positively skewed to the left. Therefore, there is a need for transformation to use linear regression.

I’ve used log10(var+1) for both variables (I used +1 for pupils who had missed 0 days school). I’m using regression because I want to add in categorical factors – gender/ethnicity etc too.

My problem is:

The audience I want to feed back to wouldn’t understand log10(y) = log(constant) + log(var2)x (and frankly neither do I).

My questions are:

a) Are there better ways of interpreting transformed variables in regression? I.e. for ever 1 day missed in 2010 they will miss 2 days in 2011 as opposed to for ever 1 log unit change in 2010 there will be x log units change in 2011?

b) Specifically, given the quoted passage from this source as follows:

“This is the negative binomial regression estimate for a one unit
increase in math standardized test score, given the other variables
are held constant in the model. If a student were to increase her
mathnce test score by one point, the difference in the logs of
expected counts would be expected to decrease by 0.0016 unit, while
holding the other variables in the model constant.”

I would like to know:

  • Is this passage saying that for every one unit increase in the score of the UNTRANSFORMED variable math leads to a 0.0016 decrease from the constant (a), so if UNTRANSFORMED maths score goes up by two points, I subtract 0.0016*2 from the constant a?
  • Does that mean that I get the geometric mean by using exponential(a)) and exponential(a+beta*2) and, that I need to calculate the percentage difference between these two to say what effect the predictor variable(s) has/have on the dependent variable?
  • Or have I got that totally wrong?

I’m using SPSS v20. Sorry for framing this in a long question.

How to interpret 礼にならない and ~にならない in general

Conversation after B saved A, B bought some drinks

A: それじゃ、助けてもらった礼にならない。

well, this is totally not gratitude for saving me (= i will do more to thank you?)

B: べつに、お礼して欲しくて、助けた訳じゃないし。

it’s not like want to you show gratitude, i didn’t really save you

a long while later after A and B part ways, A reflects on the entire experience. A did not do anything special for B in particular while they were together, they just talked about their interests.


(My interaction with her) did not show gratitude at all, but …

I am confused because normally you would want to show 礼 after being saved,but the speaker is “denying 礼” , so that this my roundabout interpretation but it is probably incorrect.

After reading these:

よく色々と億劫にならないな。- It’s amazing how he doesn’t get annoyed at all that

提出に遅れた場合は減点にならない – In the event (case) of a late submission there will not be a point deduction.

I feel like i don’t have a grasp of the distinction between ~にならない as the negative of ~になる and the ~にならない used in things like 話にならない, 洒落にならない, 問題にならない.

So when i read 礼にならない, i am not sure what is actually being implied until i read B’s reply.

Even then, “助けてもらった礼にならない” and “ちっともお礼にならなかったが” still feels very awkward for me.

Thank you for any clarifications